Marcelo Mattar

Marcelo Mattar

@marcelomattar.bsky.social

Assistant professor at NYU.

2.3K followers420 following67 posts

Super excited about this work, led by @haoliu4391.bsky.social and @s-michelmann.bsky.social. Also grateful for the many insightful conversations in which Sebastian gradually convinced me that memory encoding requires retrieval—leveraging existing semantic knowledge to regenerate the experience.

Sebastian Michelmann@s-michelmann.bsky.social

1/ Our brains probably don’t store a new copy of “red” when we encounter something red. Yet we can reinstate distinct memories of red objects. How do we reuse what we already know when encoding memories? Excited to share new work led by @haoliu4391.bsky.social, with @marcelomattar.bsky.social 🧠🧵

The dataset from our Reason to Play paper is now available on Hugging Face! Also, make sure to check out Csabi's awesome video explaining our approach.

Botos Csabi@botoscsabi.bsky.social

[Hugging Face release] We’re releasing the dataset behind Reason to Play! I’m incredibly proud of our team for reaching this milestone.

Marcelo Mattar reposted

📣 My lab at NYU currently has two open post-doc positions. We are interested in understanding perception, thinking, and hallucinations. We employ a range of cutting-edge techniques in human neuroscience (e.g., 7T fMRI, MEG, iEEG/DES) and computational approaches (e.g., DNN modeling).

Marcelo Mattar reposted

New paper! @samiyousif.bsky.social and I opine on Cognitive Maps as a Medium for Thought....now out in @cp-trendscognsci.bsky.social We discuss what cog maps are and how they might interact with a Language of Thought. This was a fun one, spanning philosophy, cog sci, and rodent neurophysiology!

Sami Yousif@samiyousif.bsky.social

Excited about a new paper (w @russellepstein.bsky.social): "Cognitive maps as a medium for thought", in @cp-trendscognsci.bsky.social! We discuss how maps are an important 'middle ground' rep. medium, and how they may interact with a 'language of thought'. www.sciencedirect.com/science/arti...

Marcelo Mattar reposted

Marcelo Mattar reposted

Marcelo Mattar reposted

Looking to work with a postdoc. Prior experience with fMRI and computational modeling required, interest in psychiatry desired. Fellowship provides three years salary plus independent travel/equipment funds. Send CV and 2-3 manuscripts (preprints OK).

Marcelo Mattar reposted

Made this overview of the intellectual history leading up to today's LLMs for my course. Any "must-have" stepping stones I missed? #NeuroAI #compneurosky #neuroskyence

Marcelo Mattar reposted

Marcelo Mattar reposted

During sleep, neurons in the hippocampus 'replay' past activity patterns. This is thought to underlie memory consolidation 🧠 After a busy day, how does the brain sort out what is worth replaying? 🤔 Check out this preprint 📃 from Tirole, Duvelle and Bendor for answers! doi.org/10.64898/202...

doi.orgTime, but not reward, shapes replay-based episodic prioritizationWhy are some experiences remembered better than others? Leading theories propose that hippocampal replay prioritizes memories for consolidation according to their expected future value. We recorded hi...

Marcelo Mattar reposted

New preprint! We show low arousal states promote hippocampal ripple genesis in sleep and wake. This bridges rodent work, where ripples predominate during sleep, with human studies reporting ripples during active behavior, identifying low arousal as a common mechanism www.biorxiv.org/content/10.6...

biorxiv.orgArousal state modulates human hippocampal ripplesHippocampal ripples are transient, high-frequency oscillations linked to memory replay and consolidation. Ripples are well-characterized in rodents to occur during periods of behavioral inactivity (i....

Marcelo Mattar reposted

A *super* exciting new paper from @lizsiefert.bsky.social!! Human hippocampal ripple rates track very closely with several indicators of low arousal, during both sleep and wake. This is probably the most compelling data I have ever been associated with.

Liz Siefert@lizsiefert.bsky.social

New preprint! We show low arousal states promote hippocampal ripple genesis in sleep and wake. This bridges rodent work, where ripples predominate during sleep, with human studies reporting ripples during active behavior, identifying low arousal as a common mechanism www.biorxiv.org/content/10.6...

Marcelo Mattar reposted

1/ 🚨 New preprint: "Closing the Loop to Discover Psychological Theories with an Automated Cognitive Scientist" Introducing AutoCog 🤖 — a fully autonomous AI system that runs the entire scientific discovery cycle in cognitive science to surface novel theories of human behavior 🧵

Check out @jonathannicholas.bsky.social 's new preprint measuring eye movements to infer individual memory retrievals during decision making, plus a cool RNN modelof that process. Can't express how excited I am about this new technique!

Jonathan Nicholas@jonathannicholas.bsky.social

We make flexible choices in new situations by knitting together information from separate relevant memories. But what governs which memories are retrieved and when? In a new preprint, we captured how people build decision variables from different memories by tracking their gaze on a blank screen.

Marcelo Mattar reposted

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... 🧵 : 👇

cell.comPerceptual multistability: a multifaceted window into brain dysfunctionsPerceptual multistability, observed across species and sensory modalities, offers valuable insights into numerous cognitive functions and dysfunctions. For instance, differences in temporal dynamics a...

Marcelo Mattar reposted

Only one week left until our PhD application deadline @mpicybernetics.bsky.social. If you’re interested in the intersection of RL, RNNs, and analyses of neural & behavioral data from novel, cross-species foraging experiments, make sure to submit your application by June 15! #NeuroJobs More info👇

Roxana Zeraati@roxana-zeraati.bsky.social

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/index.php/s/... More information about our research: www.kyb.tuebingen.mpg.de/906930/natur... #NeuroJobs

Marcelo Mattar reposted

New preprint w/ @fredcallaway.bsky.social! How does the brain decide which computations to run? We combine rational meta-reasoning with a meta-learning algorithm to build a recurrent network that learns to select computations. www.biorxiv.org/content/10.6...

biorxiv.orgLearning to select computations in recurrent neural circuitsTwo hallmarks of biological computation are its flexibility and efficiency. These features are often attributed to cognitive control processes that balance external utility against computational cost. However, how the brain could implement such adaptive control remains unknown. Here, we provide one possible answer by combining the computational theory of rational meta-reasoning with a meta-learning algorithm recently proposed as a model of prefrontal cortex. This yields a recurrent neural network model that learns to select computations. In simple choice tasks, the model approximates the algorithms and representations of optimal symbolic models and reproduces neural dynamics observed in macaque orbitofrontal cortex. In multi-step planning tasks, the model replicates key behavioral signatures of human planning strategies and captures human neural dynamics associated with step-by-step mental simulation. Our framework unifies meta-reasoning and meta-learning by showing that learning to reason can be understood as learning to learn from information generated by one’s own cognitive operations, providing a mechanistic account of how adaptive control of thought can be implemented in neural systems. ### Competing Interest Statement The authors have declared no competing interest.

Super proud to have contributed to this amazing team effort, led by @sreejan.bsky.social and @botoscsabi.bsky.social! We asked: Which AI models learn to play video games like humans, comparing both behavior and internal representations. The answer surprised us! Check out our paper and post below

Botos Csabi@botoscsabi.bsky.social

Jeee 🐦‍⬛ I am very proud of our joint effort with @sreejan.bsky.social on the project "Reason to Play" LRMs show human-like rule discovery, and their hidden states predict human brain activity during gameplay 10x better than previous methods Interactive demo + paper: botcs.github.io/reason-to-pl...

Marcelo Mattar reposted

Jeee 🐦‍⬛ I am very proud of our joint effort with @sreejan.bsky.social on the project "Reason to Play" LRMs show human-like rule discovery, and their hidden states predict human brain activity during gameplay 10x better than previous methods Interactive demo + paper: botcs.github.io/reason-to-pl...

botcs.github.ioReason to Play - Behavioral and Brain Alignment32 fMRI-scanned humans and 8 frontier open weight LLMs play ARC-AGI like games with no rules given. The reasoning models match the human learning trajectories and their hidden states predict human bra...

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.

annualreviews.orgPlanning in the Brain: It's Not What You Think It IsThe 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...

Marcelo Mattar reposted

Marcelo Mattar reposted

News! I've joined the Astera Institute to lead its neuroscience based AGI research. Backed by $1B+ commitment over the coming decade, my team will explore novel, brain-inspired architectures and algos toward safe, efficient human-like AGI, working alongside Doris Tsao. 1/ astera.org/dileep-georg...

astera.orgDileep George joins Astera to lead its neuro-inspired AGI effortDileep George is joining Astera as Head of AI, leading our AGI research division. Working alongside our Chief Scientist Doris Tsao, he and the team will explore novel, brain-inspired computational arc...

Marcelo Mattar reposted

Excited to launch Principia, a nonprofit research organisation at the intersection of deep learning theory and AI safety. Our goal is to develop theory for modern machine learning systems that can help us understand complex network behaviors, including those critical for AI safety and alignment. 1

Marcelo Mattar reposted

Marcelo Mattar reposted

Thrilled that my paper is out in the @nature.com. We explored how the brain builds complex tasks by compositionally combining simpler sub-task representations. The brain flexibly performs multiple tasks by dynamically reusing neural subspaces for sensory inputs and motor actions rdcu.be/eRVUk

rdcu.beBuilding compositional tasks with shared neural subspacesNature - The brain can flexibly perform multiple tasks by compositionally combining task-relevant neural representations.

Amazing opportunity!

Sam Gershman@gershbrain.bsky.social

If you work at the intersection of computational neuroscience and machine learning, consider applying for this postdoc position (January 2027 start date): academicpositions.harvard.edu/postings/15868 An opportunity to work with a great group of people across Harvard, MIT, and UC Berkeley.

Amazing postdoc opportunity!

Joao Barbosa@jbarbosa.org

Postdoc position in Paris: come help develop new generation human brain computer interfaces ⚡🧠💻 Interested? Contact me if you have experience with machine learning (e.g. simulation-based inference, RL, generative/diffusion models) or dynamical systems. See below for + details and retweet 🙏

Marcelo Mattar reposted

Marcelo Mattar reposted

How terrific! Free in person summer school for future PhDs who want more quantitative training before grad school; happening at one of the best places in the world to pursue it (the Gatsby at UCL).

📢 Applications open on 19 Jan for the 7-week #Mathematics #SummerSchool in London. You will develop the maths skills and intuition necessary to enter the #TheoreticalNeuroscience / #MachineLearning field. Find out more & register for the information webinar 👉 www.ucl.ac.uk/life-science...

Marcelo Mattar reposted

New preprint from the lab! A frontal motor circuit for economic decisions and actions: www.biorxiv.org/content/10.6... We use a novel task in mice, cortex-wide imaging and optogenetics, NPX recording and modelling to reveal a circuit mechanism that transforms abstract decisions to spatial actions.

biorxiv.orgA frontal motor circuit for economic decisions and actionsFlexible behaviour requires transforming abstract cognitive representations, such as value preferences, into concrete motor actions. During economic decision-making, individuals evaluate options to gu...