Excited to share our review in @cp-neuron.bsky.social with @lauriebayet.bsky.social and @mickbonner.bsky.social! We describe how implementing principles from child development can advance the mechanistic plausibility and capacities of AI models We packed A LOT into this review, here's a quick 🧵

Neuron@cp-neuron.bsky.social · 2w ago

1/5 Over a decade of comparing deep neural networks to the human brain—but what have we actually learned? Our new @cp-trendscognsci.bsky.social Feature Review synthesizes a decade of brain–DNN comparisons, asking what they reveal about brain function across vision and language.

I am *thrilled* to share this review in @currentbiology.bsky.social out today, by yours truly, @myoo.bsky.social, @michalk.bsky.social, Taro Toyoizumi, @tyrellturing.bsky.social, @taylorwwebb.bsky.social, & @hakwan.bsky.social www.sciencedirect.com/science/arti... 🧵👇

Looking to the brain to improve energy efficiency of AI

Modern artificial intelligence (AI) systems have achieved remarkable capabilities, but at an extraordinary energy cost. Training and running large-sca…

sciencedirect.com

Alpha rhythm determines speech perception rate? And speeding it up makes unintelligibly fast speech comprehensible? This is nuts! www.pnas.org/doi/abs/10.1...

Out today on @nature.com! Connecting neural specialization, population geometry, and their functional implications across the cortical hierarchy. So grateful for this beautiful and fun collaboration with @shuqiw.bsky.social, S Muscinelli, L Paninski, and @stefanofusi.bsky.social!

Rarely categorical, highly separable representations along the cortical hierarchy - Nature

Cortical circuits prioritize diversity over categorical structure, supporting a computational regime geared towards high-dimensional, highly separable neural representations.

nature.com

Excited to share that our paper has been accepted for a talk at #CogSci2026: Path Integration and Object-Location Binding Emerge in an Action-Conditioned Predictive Sequence Network Linda Ariel Ventura, Victoria Bosch, Tim C. Kietzmann, and Sushrut Thorat. Preprint: arxiv.org/abs/2602.03490. ⛓️

Path Integration and Object-Location Binding Emerge in an Action-Conditioned Predictive Sequence Network

Adaptive cognition requires structured internal models of objects and their relations. Predictive neural networks are often proposed to learn such world models, but how these are instantiated and how ...

arxiv.org

🎉 This is finally out on @natcomms.nature.com @natureportfolio.nature.com! Ever wondered what happens to your #brain when you become an expert at something? 🧠 If the answer is “yes” (+bonus points if you like #chess), check out our latest #fMRI work below! www.nature.com/articles/s41...

Andrea Costantino@costantinoai.bsky.social · 9mo ago

Super excited to share a new preprint! We asked a simple-but-big question: What changes in the brain when someone becomes an expert? Using chess ♟️ + fMRI 🧠 + representational geometry & dimensionality 📈, we ask: 1️⃣ WHAT information is encoded? 2️⃣ HOW is it structured? 3️⃣ WHERE is it expressed? 1/n

#CCN2026 will feature three GAC debates: Does NeuroAI adopt suitable methods & frameworks to understand mind & brain? Do world models emerge in prediction networks? Should neural population activity explain representations or transformations? The review period is now open. 🖥️ 2026.ccneuro.org/gac

1/ 🚨 New opinion piece: "Can we automatize scientific discovery in the cognitive sciences?" We lay out a vision for a fully automated, in-silico science of the mind, where modern AI systems run every stage of the scientific discovery cycle in cognitive science 🧵 #AutomatedDiscovery #AI4Science

Our latest story is finally in preprint! 🧠✨ How does the brain solve the (hierarchical) credit assignment problem efficiently? Cell-type-specific cortical feedback holds the key, closely approximating backprop. Paper: doi.org/10.64898/202...

Cell-type-specific cortical feedback coordinates hierarchical credit assignment

Learning is thought to arise from synaptic modifications embedded in brain-wide circuits, yet how such circuits coordinate plasticity to support complex behaviour is not known. Inspired by deep learni...

doi.org

Now published in open access! Your one-stop shop for the philosophy of language models. It's the spiritual descendant of our two-part preprint from 2024, fully updated. This should be particularly useful for anyone looking for an entry point into this rapidly growing field.

The Philosophy of Language Models

The success of large language models (LLMs) across many domains of AI research has generated intense debate. Some attribute their impressive performance on complex tasks to human-like linguistic and ...

compass.onlinelibrary.wiley.com

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 preprint!🚨 We know that LM representations can be used to predict brain responses to language. But what *features* of these representations underlie this alignment? We use SAEs to find out!

Computational models are a key part of science but discovering new ones is hard! DataDIVER discovers concise models from data, which surface new mechanistic ideas and clear predictions for future experiments From Google Deepmind Neuroscience Lab + collaborators www.biorxiv.org/content/10.6...

Our NeuroAI study made it onto the cover of Nature Machine Intelligence ❤️. In it, we demonstrate that a developmentally-inspired visual diet can drastically improve the robustness of ANN-based vision systems. www.nature.com/articles/s42... open access, open code, open weights, open science.

In a sense, memory may be the dark matter of active vision. Understanding the world through iterative glimpses requires memory. Our results indicate that the visual system already tailors each glimpse to the computational demands of that memory scaffold. /8

Exciting new work by Philip! Counterintuitively (perhaps), ‘easier’ image patches receive longer fixations during naturalistic vision…

Philip Sulewski@psulewski.bsky.social · 2mo ago

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. 🧵

There's one week left to apply to become this Fall's BCS Rising Star speaker 🌟 Please make sure to apply if you're a postdoc in the Brain & Cognitive Sciences🧠 Reposts are appreciated! 🪐

Lynn Sörensen@lynnkasorensen.bsky.social · 5mo ago

Are you a postdoc in the Brain & Cognitive Sciences? 🧠 Don't miss this opportunity to showcase your work at MIT - apply by May 31! 🌟

Today I will present our work on CorText and how to fuse neural data with LLMs in the MedARC Journal Club! I’m thankful for the invitation and looking forward! 🧠🌸 Join online: 2:15 PM UTC meet.google.com/bof-ikcz-ygh

I'm proud to say we are releasing LAION-fMRI, a densely sampled 7T fMRI dataset of natural images, with very broad stimulus sampling for testing countless hypotheses and for deeply exploring brain representations. The dataset is now available at laion-fmri.hebartlab.com What does LAION-fMRI offer? 🧵

LAION-fMRI - a 7T fMRI dataset of human vision

LAION-fMRI (LfMRI / LAION MRI dataset): 5 subjects, 25,052 launch-release natural images, 165 acquired 7T fMRI sessions with single-trial GLMsingle betas, retinotopy, localizers, and diffusion.

laion-fmri.hebartlab.com

World models in natural and artificial intelligence royalsocietypublishing.org/rsta/issue/3... - super interesting and very timely collection of articles on what it means to understand the world...

Volume 384 Issue 2320 | Philosophical Transactions of the Royal Society A | The Royal Society

Influential themed journal issues across the physical mathematical and engineering sciences.

royalsocietypublishing.org

But can we find a single time-point that offers a high-accuracy stable categorical readout from IT? No. Category information in IT can be decoded much better using a recurrent neural network with access to the whole spatiotemporal trajectory, compared to a pure ‘spatial’ code. /6

Check it out, great new work on inter-area dynamics in visual cortex! 🌀

Daniel Anthes@anthesdaniel.bsky.social · 3mo ago

Excited about our new preprint: “The illusory simplicity of the feedforward pass: evidence for the dynamical nature of stimulus encoding along the primate ventral stream” arxiv.org/abs/2604.12825 Work with Sushrut Thorat, Anna Mitola, Paolo Papale, Peter König & Tim Kietzmann 🧵 thread below

We wrote a book! www.cambridge.org/core/books/s... Consciousness science is a fascinating but bewildering field: many competing theories, little consensus, and big open questions. If you are looking for an accessible guide through this complex landscape, this book is for you.

Scientific Theories of Consciousness

Cambridge Core - Neurosciences - Scientific Theories of Consciousness

cambridge.org

OPEN POSTDOC position (part of @erc.europa.eu Consolidator DYNALANG) We build math&comp models of neural dynamics using insights from formal linguistics + ML Seeking theory-driven researchers w/ interests in language, neural dynamics, & math/comp neuroscience. Apply here: tinyurl.com/55exdpse

How is uncertainty in LLMs output reflected in internal representations? In our new work (to appear at ICML 2026), we show that the shape of internal token trajectories provides a direct geometric link to behavioral uncertainty (output entropy). 🧵(1/n)

Happy to announce the 3rd iteration of NEAT (Neuro-AI-Talks), which will take place in Osnabrück September 14th-15th 2026. NEAT is a (deliberately small scale) NeuroAI workshop that brings together researchers from neuroscience and AI. www.kietzmannlab.org/neat2026/ More information below 👇

NEAT 2026

kietzmannlab.org

Foundation models in neuroscience predict brain activity at unprecedented accuracy. But prediction ≠ understanding, and we should avoid conflating the two. New essay now out:

The Imitation Game

Foundation models in neuroscience: representational alignment versus mechanistic understanding

open.substack.com

In this beautiful review Marcel Sayre et al. give us an overview on the evolutionary origin of spatial representation incl. heading coding. @stanley-heinze.bsky.social www.sciencedirect.com/science/arti...

Head direction and the evolutionary origins of spatial representation

Spatial representations are a fundamental aspect of cognition. It remains largely unknown when and why the capacity to neurally represent space first …

sciencedirect.com

New peer-reviewed paper w/ @mheilbron.bsky.social, @predictivebrain.bsky.social & Jakub Szewczyk! Pre-onset brain encoding has been taken as evidence that brains–like LLMs–predict upcoming words. We show that the same signatures arise in systems that cannot predict. (elifesciences.org) (1/8)