@CompNeuro_lab

In Computational Machinery of Cognition Unit we study the principles of adaptive decision-making across species | Co-PIs: @roxana_zeraati and @neuroprinciples

cmc-unit.org
Joined August 2024
  • Tweets56
  • Following49
  • Followers189
2Media
Some incredibly exciting news that still feels a bit surreal: I am starting my research group @MPICybernetics! The best part? We are hiring at all levels! Curious about the science we plan to tackle, read along :)
5
13
48
2,753
We will also soon have an opening for a postdoc position. For informal inquiries, you can sign up here: forms.gle/GTWuReh4A7vucmM76
2
9
690
If these research directions resonate with you and you're interested in joining our team, we now have 2 open PhD positions! More information about the positions: nextcloud.tuebingen.mpg.de/i… More information about our research: kyb.tuebingen.mpg.de/906930/…
1
4
10
553
Of course, tackling these interdisciplinary questions requires a village! And we are incredibly lucky to have a fantastic network of collaborators to make this vision a reality, including Peter Dayan, @neuroprinciples, @XindiPoo, @lndriscoll, RoLi Lab & more!
1
2
7
344
To do this, we develop data-driven, species-specific models by integrating: 💻 🤖 computational methods (RL, dynamical systems, AI) & 🪱🐟🐀🐒👤 behavioral & neural data (from nematodes to primates) to uncover shared or species-specific mechanisms of adaptive decisions.
1
2
6
369
Our main goal is to understand how animals learn the dynamics of their environment across timescales to adapt their decisions. We study this in the context of foraging behavior, which allows us to link neural, behavioral, and environmental dynamics in naturalistic settings.
1
2
7
284
Now let’s get to the science 🪱🐟🐀🐒👤🤖
1
2
7
261
I also owe a very special thanks to @neuroprinciples who has helped me navigate every single step, discussing science, mapping out the hidden rules of academia, and teaching me when and how to pivot!
1
1
6
243
Before talking science, there is no doubt that none of this would be possible without the immense support and trust of my mentors @SelfOrgAnna, @EngelTatiana & Peter Dayan, and all my labmates! I'm so grateful for their endless support and for inspiring me every day!
1
2
7
317
Thrilled to see the first preprint of the lab out 🤩 Check it out if you need to compare dynamics in your data and RNN (or any other combinations of dynamical systems)! Code is also already available if you want to try it on your data/RNN/...: github.com/CMC-lab/fastDSA

GitHub - CMC-unit/fastDSA: Fast DSA package

Fast DSA package. Contribute to CMC-unit/fastDSA development by creating an account on GitHub.

github.com
Arman Behrad@ArmanBehradJan 8
Wanna compare dynamics across neural data, RNNs, or dynamical systems? We got a fast and furious method🏎️ The 1st preprint of my PhD 🥳 fast dynamical similarity analysis (fastDSA): 📜: arxiv.org/abs/2511.22828 💻: github.com/CMC-lab/fastDSA I’ll be @CosyneMeeting - happy to chat 😉
3
14
1,043
Wanna compare dynamics across neural data, RNNs, or dynamical systems? We got a fast and furious method🏎️ The 1st preprint of my PhD 🥳 fast dynamical similarity analysis (fastDSA): 📜: arxiv.org/abs/2511.22828 💻: github.com/CMC-lab/fastDSA I’ll be @CosyneMeeting - happy to chat 😉
1
3
2
11
1,899
Brain-like features in homeostatic deep reinforcement learning agents (Thank you @neuroprinciples!) abstracts.g-node.org/confere…
2
5
17
1,458
Replying to @StphTphsn1 @taherehtoosi
Let me (mis)use @taherehtoosi's tweet on her great study, to also draw attention to our approach to bistable perception based on reinforcement learning and decision-making models (which I think brings a complementary perspective): xitter.cf/neuroprinciples/status…
Shervin Safavi@neuroprinciples29 Aug 2022
Happy to share my first post-doc paper with Peter Dayan just published at @NeuroCellPress "Multistability, Perceptual Value, and Internal Foraging" Paper (free access before Oct 18th): authors.elsevier.com/a/1ffq3… No access? Email me at [email protected] 🧵👇
1
4
11
511
Happy to share my first post-doc paper with Peter Dayan just published at @NeuroCellPress "Multistability, Perceptual Value, and Internal Foraging" Paper (free access before Oct 18th): authors.elsevier.com/a/1ffq3… No access? Email me at [email protected] 🧵👇
13
69
6
385
🚨We believe this is a major step forward in how we study hippocampus function in healthy humans. Using novel behavioral tasks, fMRI, RL & RNN modeling, and transcranial ultrasound stimulation (TUS), we demonstrate the causal role of hippocampus in relational structure learning.
https://www.biorxiv.org/content/10.1101/2025.08.19.664920v1.full

ALT https://www.biorxiv.org/content/10.1101/2025.08.19.664920v1.full

bioRxiv Neuroscience@biorxiv_neursci19 Aug 2025
Causal necessity of human hippocampus for structure-based inference in learning biorxiv.org/content/10.1101/… #biorxiv_neursci

Causal necessity of human hippocampus for structure-based inference in learning

When meeting new individuals or encountering known individuals in new circumstances, we intuitively map out their relationships – not merely by direct experience, but by quickly inferring new...

biorxiv.org
5
33
3
122
12,590
CMC Unit retweeted
CLaE
@leafs_s2 Sep 2025
Brain-wide organization of intrinsic timescales at single-neuron resolution biorxiv.org/content/10.1101/…

Brain-wide organization of intrinsic timescales at single-neuron resolution

Variations in intrinsic neural timescales across the mammalian forebrain reflect the anatomical structure and functional specialization of brain areas and individual neurons. Yet, the organization of...

biorxiv.org
2
14
58
5,375
Still curious about uncovering internal state — completely unsupervised? We start in 30 minutes.
Arman Behrad@ArmanBehrad13 Aug 2025
#CCN2025, Poster C160: We present MoDSA – a fully unsupervised method for identifying behaviorally relevant neural states in both biological & artificial systems. From macaque V4 recordings to deep RL agents, MoDSA robustly detects state transitions without labels. See You there!
2
1
2
939
#CCN2025, Poster C160: We present MoDSA – a fully unsupervised method for identifying behaviorally relevant neural states in both biological & artificial systems. From macaque V4 recordings to deep RL agents, MoDSA robustly detects state transitions without labels. See You there!
3
1
8
1,156
In our Learning Club @CompNeuro_lab tomorrow (= June 26th, Thursday, 2 PM CET), David Clark will tell us about his recent @PNASNews paper (pnas.org/doi/10.1073/pnas.24…). If you'd like to attend, send an empty email to [email protected] to get the link!

Structure of activity in multiregion recurrent neural networks | PNAS

Neural circuits comprise multiple interconnected regions, each with complex dynamics. The interplay between local and global activity is thought to...

pnas.org
3
14
990
If you are interested in our RL models of internal cognitive processes (continuation of xitter.cf/neuroprinciples/status… and xitter.cf/neuroprinciples/status…) visit our poster "A reinforcement learning model of task-modulated perception" today!
Shervin Safavi@neuroprinciples13 Dec 2024
Happy to share my second paper with Peter Dayan on our decision-theoretic approach to perceptual multistability (see the tweeprint of the first paper here xitter.cf/neuroprinciples/status…): A decision-theoretic model of multistability biorxiv.org/content/10.1101/…
1
2
20
1,193
Load more