Excited to share that my first-ever first-author paper (w/ @atabk.bsky.social @AngeliqueDelarazan @zreagh.bsky.social) is now out in PNAS! Building a story to link two objects boosts associative memory and inference, but not memory for the objects themselves. www.pnas.org/doi/10.1073/... 🧵

Active linking through narratives facilitates associative inference | PNAS

In daily life, we often draw inferences about novel associations from prior experiences. This ability, associative inference, is thought to be a ke...

pnas.org

Can MEG provide a non-invasive window onto human ripple physiology during sleep? Our new preprint suggests it can. Led by the fantastic @fabian31415.bsky.social. Many thanks to our collaborators @olejensen.bsky.social , @doellerlab.bsky.social , and @tobiasstaudigl.bsky.social #Neuroskyence

Fabian Schwimmbeck@fabian31415.bsky.social · 2w ago

Can we study MTL ripples non-invasively? New preprint! We show that source-resolved MEG tracks ripple-like activity during sleep that: • localizes to the MTL • follows the canonical SO–spindle hierarchy • matches an iEEG benchmark • indexes memory reactivation www.biorxiv.org/content/10.6...

This is a really interesting and important study: www.nature.com/articles/s41... However, isn't it concerning that sweeping conclusions about functional organization of cortex are being reached on the basis of a single task?

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

Thanks to Kshipra Guranandan, my notebook on how to design efficient fMRI experiments (for activity, connectivity and pattern analysis) is now available in Python as a Jupyter notebook (as well as previous Matlab): github.com/RikHenson/fM...

GitHub - RikHenson/fMRIefficiency

Contribute to RikHenson/fMRIefficiency development by creating an account on GitHub.

github.com

Memory palace like strategy can naturally emerge in an RNN trained to do the free recall task! Very honored to be involved in this work. Congrats again @moufanli.bsky.social :)

Moufan Li@moufanli.bsky.social · 2w ago

Our paper is out in @natmachintell.nature.com! We trained multiple RNNs to perform free recall. The best-performing ones learned a strategy akin to the memory palace technique. See thread below for more info. www.nature.com/articles/s42...

New preprint from the lab! “A number simplex in the human medial temporal lobe” led by @hanlin-zhu.bsky.social 🧵 www.biorxiv.org/content/10.6...

A number simplex in the human medial temporal lobe

Humans handle numbers nimbly, suggesting a richer neural manifold structure than the prevalent mental number line model. In populations of medial temporal lobe (MTL) neurons in humans performing two s...

biorxiv.org

From a great collaboration with @lorenzoposani.com, @shuqiw.bsky.social, Samuel Muscinelli, Liam Paninski, now in Nature: www.nature.com/articles/s41...

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

Hippocampal engrams configure prefrontal context representations to guide flexible decisions Impressive combo of methods to manipulate and record this process: www.biorxiv.org/content/10.6...

Hippocampal engrams configure prefrontal context representations to guide flexible decisions

Flexible behavior requires using past experiences to configure cortical computations to suit current task demands. A central question in neuroscience is how memory representations control such reconfi...

biorxiv.org

Provocative title ftw. (Tho it inspired @kristorpjensen.bsky.social to give a talk titled "planning in the brain: it's not what nathaniel thinks it is")

Annual Reviews@annualreviews.bsky.social · 3w ago

The 2026 volume of the Annual Review of Neuroscience is now online 🧠 The most read article is "Planning in the Brain: It's Not What You Think It Is" by @marcelomattar.bsky.social and @nathanieldaw.bsky.social

Our paper "Neural timescales from a computational perspective" is finally out in @natneuro.nature.com: www.nature.com/articles/s41... We discuss how computational models and methods can distill empirical observations on neural timescales into quantitative, testable theories of brain computation.

Neural timescales from a computational perspective - Nature Neuroscience

This review integrates computational approaches to provide a unified view on how data analysis methods, biophysical mechanistic models and machine learning approaches can help to uncover the origins a...

nature.com

We're finally getting the tools to study replay in humans. But why bother? What can we learn from it that we don't know from rodent recordings? Our thoughts are now out in a new TICS Forum Great fun writing this w @smfleming.bsky.social + @maritpetzka.bsky.social authors.elsevier.com/sd/article/S...

What can we learn from studying replay in humans?

Noninvasive methods are making it possible to study neural replay in humans. Although focused on alignment with rodent findings, human research often …

authors.elsevier.com

I am so excited about this finding from @pengrui-han.bsky.social and @andreadevarda.bsky.social, also with Jacob Andreas! Perhaps modularity is inevitable in intelligent systems, biological or in silico. :)

pengrui-han.bsky.social@pengrui-han.bsky.social · last mo.

The human brain is strikingly modular: distinct networks for language, formal reasoning, social reasoning, physical reasoning. Is this fundamental to intelligent systems, or an accident of evolution? In our new preprint, we find the same modular organization emerges in LLMs.

🧠 New preprint! How well can current algorithms *actually* detect neural "replay" in the human brain under absolutely optimal conditions? We built FASTIMAGES: A combined MEG + fMRI benchmark with KNOWN neural sequences, so replay-detection methods can finally be validated against a ground truth.

FASTIMAGES: Validating replay detection methods in human neuroimaging

A combined MEG + fMRI benchmark dataset with known neural sequences to validate replay detection methods (TDLM and SODA).

cimh-clinical-psychology.github.io

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:

Human learning of noninvasive brain–computer interfaces via manifold geometry - Nature Neuroscience

Busch et al. use nonlinear neural manifolds to help humans gain rapid control over a noninvasive brain–computer interface, allowing them to learn how to play a video game with real-time fMRI neurofeed...

nature.com

"Prior knowledge biases episodic memory by filling in the gaps in imprecise memories: Simulating aging effects in younger adults" New paper from @michelleramey.bsky.social Read more here: www.sciencedirect.com/science/arti...

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

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... 🧵 : 👇

Perceptual multistability: a multifaceted window into brain dysfunctions

Perceptual multistability, observed across species and sensory modalities, offers valuable insights into numerous cognitive functions and dysfunctions. For instance, differences in temporal dynamics a...

cell.com

Trends in Cognitive Sciences@cp-trendscognsci.bsky.social · 4mo ago

Online Now: Perceptual multistability: a multifaceted window into brain dysfunctions

Feeling really proud of this one: a review of Bayesian efficient coding with Máté Lengyel and Paul Bays (@bayslab.org) is now available online in TICS! authors.elsevier.com/a/1nF3i4sIRv...

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.

Memorization vs. generalization in deep learning: implicit biases, benign overfitting, and more

Or: how I learned to stop worrying and love the memorization

infinitefaculty.substack.com

Super excited by this manuscript led by the amazing @brissend.bsky.social By combining fMRI, TMS, and modeling, we finally have causal evidence that the cerebellum is contributing to brain-wide working memory representations and recall performance! 1/n

bioRxiv Neuroscience@biorxiv-neursci.bsky.social · 3mo ago

Cerebellar perturbation impairs human working memory and degrades spatial tuning throughout cortex https://www.biorxiv.org/content/10.64898/2026.05.14.724968v1

🎉 WE’RE OFFICIALLY ANNOUNCING FOR 2026 🎉 On June 20, we’re bringing researchers, technologists, artists, economists, cognitive scientists, founders, academics, writers, and curious humans together in Washington, DC. 🧵

🚨 New Preprint 🚨 I’m excited to share the first paper from my postdoc. We found age differences in the timescales of neural activity in the hippocampus during movie viewing 👀 These timescales were related to memory specificity in an interesting way (spoiler: the hippocampus may not be special!?)

bioRxiv Neuroscience@biorxiv-neursci.bsky.social · 3mo ago

Age Differences in Hippocampal Neural Timescales During Movie-Viewing https://www.biorxiv.org/content/10.64898/2026.05.05.723065v1

THE NEURAL GEOMETRY SERIES The World Inside Neural Networks www.goodfire.ai/research/the...

The World Inside Neural Networks

How neural geometry will unlock understanding and control of AI

goodfire.ai

NEW PAPER! Here, my postdoc, Biru Dudhabhate, and I review the history of one of the most famous neuroscience hypotheses, the dopamine reward prediction error, and reflect on what made it such a major advance for the field. @uabneuro.bsky.social www.frontiersin.org/journals/com...

Frontiers | A brief history of dopamine prediction errors

Dopamine signaling has become closely associated with reward prediction errors (RPEs)–the difference between expected and experienced value. Although not wit...

frontiersin.org

Because this is a NeuroAI-flavored paper, Xinyuan and I decided to make a blogpost about it: psywalkeryanxy.github.io/posts/scienc...

Polysemanticity in Human Hippocampal Neurons – Xinyuan Yan

When single neurons encode word meanings (concepts), they use the computational principle of superposition.

psywalkeryanxy.github.io

In collaboration with @monicarosenb.bsky.social , we showed that individual diffs in LTM encoding is uniquely predicted by inter-electrode correlations even controlling for working memory abilities. This suggests that WM & LTM encoding are separate abilities coded by different neural signatures! 1/n

Imaging Neuroscience@imagingneurosci.bsky.social · 3mo ago

New paper in Imaging Neuroscience by Chong Zhao, Edward K. Vogel, and Monica D. Rosenberg: A unique neural signature of long-term memory encoding from EEG inter-electrode correlation doi.org/10.1162/IMAG...

New paper from the lab! "Plasticity and language in the anaesthetized human hippocampus" www.nature.com/articles/s41...

Plasticity and language in the anaesthetized human hippocampus - Nature

In the hippocampus, complex processing of sensory stimuli occurs even in the unconscious state.

nature.com

The final typeset version of this paper is online now! Check out the new analysis approach that Narjes developed to track changes in event structure at multiple timescales, allowing us to see how event dynamics in a film clip change with repeated viewing

Mariam Aly@mariamaly.bsky.social · 4mo ago

How do the brain’s event representations change as we gain familiarity with an experience? Brain regions’ representations can become coarser or finer as events become familiar. Slow-timescale structure predicts memory. Excited to share this work w/ Narjes Al-Zahli & @chrisbaldassano.bsky.social!

New paper out in cognition with @arikahn.bsky.social, @nathanieldaw.bsky.social, Cate Hartley, and @katenuss.bsky.social !! We show that children 👶 use predictive representations (e.g. SR) to guide their choices, providing an account of how they can make flexible choices in a changing world

Children leverage predictive representations for flexible, value-guided choice

By harnessing a mental model of how the world works, learners can make flexible choices in changing environments. However, while children and adolesce…

sciencedirect.com

New Annual Review with @nathanieldaw.bsky.social: “Planning in the Brain: It's Not What You Think It Is.” We argue that the brain's 'planning' machinery is mostly used for learning from simulated experience, and that thinking prospectively at decision time is just one special case of this process.

Planning in the Brain: It&apos;s Not What You Think It Is

The neuroscience of planning has long been analogized to search algorithms in artificial intelligence (AI), which simulate future actions to guide immediate choices. We argue that advances in both neu...

annualreviews.org