Thrilled to finally share this work! 🧠🔊

Using a new reinforcement-free task we show mice (like humans) extract abstract structure from sound (unsupervised) & dCA1 is causally required by building factorised, orthogonal subspaces of abstract rules.

Led by Dammy Onih!
www.biorxiv.org/content/10.6...

Biological insights into schizophrenia from ancestrally diverse populations Nature - Genome-wide association studies incorporating data for populations of African ancestry provide an expanded view of the genetic basis of schizophrenia, which has previously been studied...

Now out in @nature.com: Biological insights into schizophrenia from ancestrally diverse populations.

@sinaibrain.bsky.social @sinaigenetics.bsky.social cs.bsky.social @timbigdeli.bsky.social #CDNeurogenomics #MountSinaiPsych #MillionVeteranProgram and many collaborators
Read: rdcu.be/eZ7he

ADHD is on the rise, but why? The more we learn, the less we seem to understand this condition.

Rates of ADHD have been rising quickly over the past few decades, for reasons that are not entirely clear — a mystery that underscores how much we still have to learn about the condition.

go.nature.com/49TQWG5

I’m very happy to share the latest from my lab published in @Nature

Hippocampal neurons that initially encode reward shift their tuning over the course of days to precede or predict reward.

Full text here:
rdcu.be/eY5nh

All theory is wrong until verified by data. Greatly indebted to @mhyaghoubi.bsky.social, @markbrandonlab.bsky.social, @douglasresearch.bsky.social for finding the hippocampus encoding reward prediction! Grateful to my advisor @cpehlevan.bsky.social, @kempnerinstitute.bsky.social.
#RL #hippocampus

𝗕𝗿𝗮𝗶𝗻-𝗯𝗼𝗱𝘆 𝗽𝗵𝘆𝘀𝗶𝗼𝗹𝗼𝗴𝘆:
𝗟𝗼𝗰𝗮𝗹, 𝗿𝗲𝗳𝗹𝗲𝘅, 𝗮𝗻𝗱 𝗰𝗲𝗻𝘁𝗿𝗮𝗹 𝗰𝗼𝗺𝗺𝘂𝗻𝗶𝗰𝗮𝘁𝗶𝗼𝗻
Excellent review paper about reactive and anticipatory processes.
#neuroskyence
doi.org/10.1016/j.ce...

I am extremely grateful to be awarded the National University of Singapore (NUS) Development Grant, and to be a Young NUS Fellow! Look forward to collaborating with the Yong Loo Lin School of Medicine on exciting projects. This is my first grant and hopefully many more to come! #NUS #NeuroAI

I wrote a Comment on neurotheory, and now you can read it!

Some thoughts on where neurotheory has and has not taken root within the neuroscience community, how it has shaped those subfields, and where we theorists might look next for fresh adventures.

www.nature.com/articles/s41...

🧵 New paper! We studied depression symptoms and goal-directed decisions under uncertainty

@shiyiliang.bsky.social, with @evanrussek.bsky.social & @robbrutledge.bsky.social

Surprisingly, we found that apathy–anhedonia was linked to enhanced goal-directed behavior. www.biorxiv.org/content/10.1...

Not just for AI but these theories can improve our understanding of biological networks too!

On the left is a rabbit. On the right is an elephant. But guess what: They’re the *same image*, rotated 90°!

In @currentbiology.bsky.social, @chazfirestone.bsky.social & I show how these images—known as “visual anagrams”—can help solve a longstanding problem in cognitive science. bit.ly/45BVnCZ

trying this with GPT-5 and charting new frontiers in gaslighting

Connectivity structure and dynamics of nonlinear recurrent neural networks Studies of the dynamics of nonlinear recurrent neural networks often assume independent and identically distributed couplings, but large-scale connectomics data indicate that biological neural circuit...

Wanted to share a new version (much cleaner!) of a preprint on how connectivity structure shapes collective dynamics in nonlinear RNNs. Neural circuits have highly non-iid connectivity (e.g., rapidly decaying singular values, structured singular-vector overlaps), unlike classical random RNN models.

3. We present TeDFA-δ, a bio. plaus. deep spiking RL model that leverages temporal integration and weak learning rules to outperform standard MLPs+BP for policy learning, highlighting the importance of neural dynamics over credit assignment for effective control:

2025.ccneuro.org/poster/?id=S...

2. We developed a bio. plaus. computational model of the dentate gyrus that shows how both impaired synaptic plasticity and increased neurogenesis—modulated by Cbln4-Neo1 complex—disrupt behavioral pattern separation:

2025.ccneuro.org/poster/?id=P...

1. We developed a RNN-based meta-RL framework that models schizophrenia-like decision-making deficits. We see a positive correlation between the number of dynamical attractor states and suboptimal behavior:

2025.ccneuro.org/poster/?id=4...

1 proceeding and 2 extended abstracts at Cognitive Computational Neuroscience (CCN) Conference 2025! Short summaries and links are in the thread. Look forward to the discussions! #CCN25

The emergence of NeuroAI: bridging neuroscience and artificial intelligence — a Comment article by Sadra Sadeh & Claudia Clopath

@sdrsd.bsky.social @clopathlab.bsky.social
‪
#neuroscience #neuroskyence

www.nature.com/articles/s41...

Attractor dynamics of working memory explain a concurrent evolution of stimulus-specific and decision-consistent biases in visual estimation People exhibit biases when perceiving features of the world, shaped by both external stimuli and prior decisions. By tracking behavioral, neural, and mechanistic markers of stimulus- and decision-rela...

Excited to share that our paper is now out in Neuron @cp-neuron.bsky.social (dlvr.it/TM9zJ8).

Our perception isn't a perfect mirror of the world. It's often biased by our expectations and beliefs. How do these biases unfold over time, and what shapes their trajectory? A summary thread. (1/13)

What do representations tell us about a system? Image of a mouse with a scope showing a vector of activity patterns, and a neural network with a vector of unit activity patterns Common analyses of neural representations: Encoding models (relating activity to task features) drawing of an arrow from a trace saying [on_____on____] to a neuron and spike train. Comparing models via neural predictivity: comparing two neural networks by their R^2 to mouse brain activity. RSA: assessing brain-brain or model-brain correspondence using representational dissimilarity matrices

In neuroscience, we often try to understand systems by analyzing their representations — using tools like regression or RSA. But are these analyses biased towards discovering a subset of what a system represents? If you're interested in this question, check out our new commentary! Thread:

A landmark volume, The Handbook of Dopamine, is now online:
www.sciencedirect.com/handbook/han...

Big kudos to the editors, Stephanie Cragg and Mark Walton, for putting this together.

Coming March 17, 2026!
Just got my advance copy of Emergence — a memoir about growing up in group homes and somehow ending up in neuroscience and AI. It’s personal, it’s scientific, and it’s been a wild thing to write. Grateful and excited to share it soon.

How can we test theories in neuroscience? Take a variable predicted to be important by the theory. It could fail to be observed because it's represented in some nonlinear, even distributed way. Or it could be observed but not be causal because the network is a reservoir. How can we deal with this?

This summer my lab's journal club somewhat unintentionally ended up reading papers on a theme of "more naturalistic computational neuroscience". I figured I'd share the list of papers here 🧵:

First #ICML2025 conference proceeding (icml.cc/virtual/2025...)! We (@frostedblakess.bsky.social, @jzv.bsky.social, @cpehlevan.bsky.social) developed a simple model to better understand state representation learning dynamics in both artificial and biological intelligent systems!

State representation learning in the hippocampus?

I'm heading back to Singapore for ICLR25! Hit me up for discussions or where to find good food!

#neuroai #home

Interestingly, we found no significant difference in under and over-updating behavior in Schizophrenia patient data (Nassar et al. 2021). Instead, analyzing the behavior using the delta area metric showed a significant difference, suggesting the utility of model-guided human-behavior data analysis.

We used a fixed point finder algorithm and found that suboptimal agents (lower delta area value) exhibited smaller number of unstable fixed points compared to more optimal agents. The number of stable fixed points remained consistent across the delta area metric.

Besides the (1) reward discount factor, we explored (2) prediction error scaling, (3) probability of disrupting RNN dynamics, (4) rollout buffer length. Each hyperparameter differently influenced the suboptimal decision making behavior, which we termed as delta area.