Andrea Costantino
@costantinoai.bsky.social
👀🧠🤖 cognitive neuroscientist @ Hoplab, KU Leuven | interested in vision and learning
Reposted by Andrea Costantino
Online Now: The Umwelt Representation Hypothesis: rethinking Universality
The Umwelt Representation Hypothesis: rethinking Universality
Recent studies reveal striking representational alignment between artificial neural networks (ANNs) and biological brains, leading to proposals that all sufficiently capable systems converge on universal representations of reality. We argue that this claim of Universality is premature. We introduce the Umwelt Representation Hypothesis, which proposes that alignment arises not from convergence toward a single global optimum but from overlap in the ecological constraints under which systems develop. We review empirical evidence showing that representational differences between species, individuals, and ANNs are systematic and adaptive, which is difficult to reconcile with Universality. Finally, we reframe ANN model comparison as a method for mapping clusters of alignment in the ecological constraint space rather than as a search for a single optimal world model.
dlvr.it
August 8, 2026 at 12:40 PM
Online Now: The Umwelt Representation Hypothesis: rethinking Universality
Reposted by Andrea Costantino
And come see the substack series: danyamins.substack.com/p/the-theory...
July 13, 2026 at 1:49 PM
And come see the substack series: danyamins.substack.com/p/the-theory...
Reposted by Andrea Costantino
Ever wondered what happens to your #brain when you become an expert at something? 🧠
Check it out at www.nature.com/articles/s41...!
Check it out at www.nature.com/articles/s41...!
July 9, 2026 at 12:03 PM
Ever wondered what happens to your #brain when you become an expert at something? 🧠
Check it out at www.nature.com/articles/s41...!
Check it out at www.nature.com/articles/s41...!
🎉 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...
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...
July 7, 2026 at 9:01 PM
🎉 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...
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...
Conference wrap-ups are about science. And #VSS2026 had some GREAT science.
But now that VSS is (sadly) over, what I really want to celebrate is our community. The old and new (!!) friends, the science talks at 1am, the tiki, and all the good times.
Thank you all! THIS is what I work for
But now that VSS is (sadly) over, what I really want to celebrate is our community. The old and new (!!) friends, the science talks at 1am, the tiki, and all the good times.
Thank you all! THIS is what I work for
May 21, 2026 at 12:42 AM
Conference wrap-ups are about science. And #VSS2026 had some GREAT science.
But now that VSS is (sadly) over, what I really want to celebrate is our community. The old and new (!!) friends, the science talks at 1am, the tiki, and all the good times.
Thank you all! THIS is what I work for
But now that VSS is (sadly) over, what I really want to celebrate is our community. The old and new (!!) friends, the science talks at 1am, the tiki, and all the good times.
Thank you all! THIS is what I work for
Reposted by Andrea Costantino
Do you work or study in the fields of psychology, neuroscience, computer science, artificial intelligence, or philosophy?
What does the term 'representation' mean to you?
We invite you to participate in a brief survey on key conceptual questions across fields.
eu.surveymonkey.com/r/VX9GNXM
What does the term 'representation' mean to you?
We invite you to participate in a brief survey on key conceptual questions across fields.
eu.surveymonkey.com/r/VX9GNXM
February 12, 2026 at 1:39 PM
Do you work or study in the fields of psychology, neuroscience, computer science, artificial intelligence, or philosophy?
What does the term 'representation' mean to you?
We invite you to participate in a brief survey on key conceptual questions across fields.
eu.surveymonkey.com/r/VX9GNXM
What does the term 'representation' mean to you?
We invite you to participate in a brief survey on key conceptual questions across fields.
eu.surveymonkey.com/r/VX9GNXM
Reposted by Andrea Costantino
🔥 We have a new preprint! 🔥
It's really nice, I recommend reading it
Here's a summary question: Can FFA activation predict RT to faces?
⏬ Check out the answer below ⏬
@hansopdebeeck.bsky.social
www.biorxiv.org/content/10.6...
It's really nice, I recommend reading it
Here's a summary question: Can FFA activation predict RT to faces?
⏬ Check out the answer below ⏬
@hansopdebeeck.bsky.social
www.biorxiv.org/content/10.6...
www.biorxiv.org
February 18, 2026 at 10:02 PM
🔥 We have a new preprint! 🔥
It's really nice, I recommend reading it
Here's a summary question: Can FFA activation predict RT to faces?
⏬ Check out the answer below ⏬
@hansopdebeeck.bsky.social
www.biorxiv.org/content/10.6...
It's really nice, I recommend reading it
Here's a summary question: Can FFA activation predict RT to faces?
⏬ Check out the answer below ⏬
@hansopdebeeck.bsky.social
www.biorxiv.org/content/10.6...
Reposted by Andrea Costantino
Returning home inspired after a great visit to KU Leuven, where I presented our work on CorText in the Brain & Cognition group. Thank you for the invitation and great discussions! @hansopdebeeck.bsky.social @costantinoai.bsky.social
(pictured: the magnificent architecture of the Liège station)
(pictured: the magnificent architecture of the Liège station)
February 11, 2026 at 10:38 AM
Returning home inspired after a great visit to KU Leuven, where I presented our work on CorText in the Brain & Cognition group. Thank you for the invitation and great discussions! @hansopdebeeck.bsky.social @costantinoai.bsky.social
(pictured: the magnificent architecture of the Liège station)
(pictured: the magnificent architecture of the Liège station)
Reposted by Andrea Costantino
Our paper is out in @natneuro.nature.com!
www.nature.com/articles/s41...
We develop a geometric theory of how neural populations support generalization across many tasks.
@zuckermanbrain.bsky.social
@flatironinstitute.org
@kempnerinstitute.bsky.social
1/14
www.nature.com/articles/s41...
We develop a geometric theory of how neural populations support generalization across many tasks.
@zuckermanbrain.bsky.social
@flatironinstitute.org
@kempnerinstitute.bsky.social
1/14
February 10, 2026 at 3:56 PM
Our paper is out in @natneuro.nature.com!
www.nature.com/articles/s41...
We develop a geometric theory of how neural populations support generalization across many tasks.
@zuckermanbrain.bsky.social
@flatironinstitute.org
@kempnerinstitute.bsky.social
1/14
www.nature.com/articles/s41...
We develop a geometric theory of how neural populations support generalization across many tasks.
@zuckermanbrain.bsky.social
@flatironinstitute.org
@kempnerinstitute.bsky.social
1/14
Reposted by Andrea Costantino
Can the adult brain acquire typical representations after missing an essential early-life experience? New work with Marge Livingstone: “Early face deprivation leads to long-lasting deficits in cortical face processing”.
www.biorxiv.org/content/10.6...
www.biorxiv.org/content/10.6...
www.biorxiv.org
February 6, 2026 at 10:43 PM
Can the adult brain acquire typical representations after missing an essential early-life experience? New work with Marge Livingstone: “Early face deprivation leads to long-lasting deficits in cortical face processing”.
www.biorxiv.org/content/10.6...
www.biorxiv.org/content/10.6...
Reposted by Andrea Costantino
New paper out at PNAS: www.pnas.org/doi/10.1073/...
Revisiting the high-dimensional geometry of population responses in the visual cortex with @jpillowtime.bsky.social. The review took forever because a reviewer was doubtful our new estimator can infer eigenvalues beyond the rank of the data! (1/6)
Revisiting the high-dimensional geometry of population responses in the visual cortex with @jpillowtime.bsky.social. The review took forever because a reviewer was doubtful our new estimator can infer eigenvalues beyond the rank of the data! (1/6)
PNAS
Proceedings of the National Academy of Sciences (PNAS), a peer reviewed journal of the National Academy of Sciences (NAS) - an authoritative source of high-impact, original research that broadly spans...
www.pnas.org
January 27, 2026 at 4:34 PM
New paper out at PNAS: www.pnas.org/doi/10.1073/...
Revisiting the high-dimensional geometry of population responses in the visual cortex with @jpillowtime.bsky.social. The review took forever because a reviewer was doubtful our new estimator can infer eigenvalues beyond the rank of the data! (1/6)
Revisiting the high-dimensional geometry of population responses in the visual cortex with @jpillowtime.bsky.social. The review took forever because a reviewer was doubtful our new estimator can infer eigenvalues beyond the rank of the data! (1/6)
Reposted by Andrea Costantino
This paper had a pretty shocking headline result (40% of voxels!), so I dug into it, and I think it is wrong. Essentially: they compare two noisy measures and find that about 40% of voxels have different sign between the two. I think this is just noise!
Would love to hear expert views on this paper. It appears to show that the operationalization of brain activity the field has relied on for 3 decades—the BOLD response—is not actually a sensible measure of brain activity.
www.nature.com/articles/s41...
www.nature.com/articles/s41...
BOLD signal changes can oppose oxygen metabolism across the human cortex - Nature Neuroscience
Using quantitative brain imaging, the authors show opposite fMRI BOLD signal to metabolic activity due to variable oxygen extraction across the human cortex. This questions the canonical interpretatio...
www.nature.com
January 5, 2026 at 5:22 PM
This paper had a pretty shocking headline result (40% of voxels!), so I dug into it, and I think it is wrong. Essentially: they compare two noisy measures and find that about 40% of voxels have different sign between the two. I think this is just noise!
A wonderful surprise to see our latest study on expertise so beautifully explained by @bjbalas.bsky.social on @lichess.org.
Go check it out!
Go check it out!
After a rather long delay, I've finally got a new Science of #Chess post up on lichess! This time I'm looking at a very neat #neuroscience paper (co-authored by @hansopdebeeck.bsky.social) using chess as a model for studying expertise. Enjoy! #cogneuro #PsychSciSky
Networks for expertise in the chess players' brain
Chess can be an important way of investigating how learning and expertise are manifest in the nervous system
lichess.org
January 30, 2026 at 1:49 AM
A wonderful surprise to see our latest study on expertise so beautifully explained by @bjbalas.bsky.social on @lichess.org.
Go check it out!
Go check it out!
Reposted by Andrea Costantino
Y’all are reading this paper in the wrong way.
We love to trash dominant hypothesis, but we need to look for evidence against the manifold hypothesis elsewhere:
This elegant work doesn't show neural dynamics are high D, nor that we should stop using PCA
It’s quite the opposite!
(thread)
We love to trash dominant hypothesis, but we need to look for evidence against the manifold hypothesis elsewhere:
This elegant work doesn't show neural dynamics are high D, nor that we should stop using PCA
It’s quite the opposite!
(thread)
“Our findings challenge the conventional focus on low-dimensional coding subspaces as a sufficient framework for understanding neural computations, demonstrating that dimensions previously considered task-irrelevant and accounting for little variance can have a critical role in driving behavior.”
Neural dynamics outside task-coding dimensions drive decision trajectories through transient amplification
Most behaviors involve neural dynamics in high-dimensional activity spaces. A common approach is to extract dimensions that capture task-related variability, such as those separating stimuli or choice...
www.biorxiv.org
November 25, 2025 at 4:16 PM
Y’all are reading this paper in the wrong way.
We love to trash dominant hypothesis, but we need to look for evidence against the manifold hypothesis elsewhere:
This elegant work doesn't show neural dynamics are high D, nor that we should stop using PCA
It’s quite the opposite!
(thread)
We love to trash dominant hypothesis, but we need to look for evidence against the manifold hypothesis elsewhere:
This elegant work doesn't show neural dynamics are high D, nor that we should stop using PCA
It’s quite the opposite!
(thread)
Huh, funny. 40 years passed, but this is still so relatable
November 25, 2025 at 8:28 AM
Huh, funny. 40 years passed, but this is still so relatable
Reposted by Andrea Costantino
Not directly related to @neurograce.bsky.social’s comment here but to the thread: I am baffled by the number of people who seem to think the claim is that all brain computation is low-d vs. the *emprical* finding that task-related neural activity is (linear) low-d in many (not all) cases.
This kind of stuff is why I say that I worry that the tools of neuroscience are not properly vetted
“Our findings challenge the conventional focus on low-dimensional coding subspaces as a sufficient framework for understanding neural computations, demonstrating that dimensions previously considered task-irrelevant and accounting for little variance can have a critical role in driving behavior.”
November 24, 2025 at 1:16 AM
Not directly related to @neurograce.bsky.social’s comment here but to the thread: I am baffled by the number of people who seem to think the claim is that all brain computation is low-d vs. the *emprical* finding that task-related neural activity is (linear) low-d in many (not all) cases.
Reposted by Andrea Costantino
1. 🧵 Thread: What happens to the visual brain after early transient blindness?
Our new Nature Communications paper examines a rare population: people born with dense bilateral cataracts—a short blindness occurring during a critical window of visual development.
🔗 rdcu.be/eQjMH
Our new Nature Communications paper examines a rare population: people born with dense bilateral cataracts—a short blindness occurring during a critical window of visual development.
🔗 rdcu.be/eQjMH
November 19, 2025 at 9:07 AM
1. 🧵 Thread: What happens to the visual brain after early transient blindness?
Our new Nature Communications paper examines a rare population: people born with dense bilateral cataracts—a short blindness occurring during a critical window of visual development.
🔗 rdcu.be/eQjMH
Our new Nature Communications paper examines a rare population: people born with dense bilateral cataracts—a short blindness occurring during a critical window of visual development.
🔗 rdcu.be/eQjMH
Reposted by Andrea Costantino
Low-Dimensional and Optimised Representations of High-Level Information in the Expert Brain https://www.biorxiv.org/content/10.1101/2025.11.12.688012v1
November 12, 2025 at 5:16 PM
Low-Dimensional and Optimised Representations of High-Level Information in the Expert Brain https://www.biorxiv.org/content/10.1101/2025.11.12.688012v1
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
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
November 12, 2025 at 10:56 PM
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
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
Reposted by Andrea Costantino
New paper (and thread) on the representational dynamics of the main dimensions of object space: jov.arvojournals.org/article.aspx... 1/n
Representational dynamics of the main dimensions of object space: Face/body selectivity aligns temporally with animal taxonomy but not with animacy | JOV | ARVO Journals
jov.arvojournals.org
November 9, 2025 at 1:48 PM
New paper (and thread) on the representational dynamics of the main dimensions of object space: jov.arvojournals.org/article.aspx... 1/n
Reposted by Andrea Costantino
It's finally out! Our work with @hansopdebeeck.bsky.social & @costantinoai.bsky.social is published.
We went looking for dissociations between types of recurrence in DNNs, but we found something quite different.. hopefully that can tell us somehting about our models!
rdcu.be/eLwBA
We went looking for dissociations between types of recurrence in DNNs, but we found something quite different.. hopefully that can tell us somehting about our models!
rdcu.be/eLwBA
October 27, 2025 at 4:18 PM
It's finally out! Our work with @hansopdebeeck.bsky.social & @costantinoai.bsky.social is published.
We went looking for dissociations between types of recurrence in DNNs, but we found something quite different.. hopefully that can tell us somehting about our models!
rdcu.be/eLwBA
We went looking for dissociations between types of recurrence in DNNs, but we found something quite different.. hopefully that can tell us somehting about our models!
rdcu.be/eLwBA
Reposted by Andrea Costantino
We’re hiring! We’re looking for two RAs to study neuroplasticity in sight loss, sight rescue and development in children and adults @ucl.ac.uk using a wide range of neuroimaging and behavioral methods. Please help spread the word! Apply by 16 Oct! t.ly/q3aYe #neurojobs #NeuroSkyence
Research Assistant at UCL
Recruiting now: Research Assistant on jobs.ac.uk. Click for details and explore more academic job opportunities on the top job board
t.ly
September 27, 2025 at 10:57 AM
We’re hiring! We’re looking for two RAs to study neuroplasticity in sight loss, sight rescue and development in children and adults @ucl.ac.uk using a wide range of neuroimaging and behavioral methods. Please help spread the word! Apply by 16 Oct! t.ly/q3aYe #neurojobs #NeuroSkyence
Reposted by Andrea Costantino
🧠 New preprint: Why do deep neural networks predict brain responses so well?
We find a striking dissociation: it’s not shared object recognition. Alignment is driven by sensitivity to texture-like local statistics.
📊 Study: n=57, 624k trials, 5 models doi.org/10.1101/2025...
We find a striking dissociation: it’s not shared object recognition. Alignment is driven by sensitivity to texture-like local statistics.
📊 Study: n=57, 624k trials, 5 models doi.org/10.1101/2025...
September 8, 2025 at 6:32 PM
🧠 New preprint: Why do deep neural networks predict brain responses so well?
We find a striking dissociation: it’s not shared object recognition. Alignment is driven by sensitivity to texture-like local statistics.
📊 Study: n=57, 624k trials, 5 models doi.org/10.1101/2025...
We find a striking dissociation: it’s not shared object recognition. Alignment is driven by sensitivity to texture-like local statistics.
📊 Study: n=57, 624k trials, 5 models doi.org/10.1101/2025...
It was great reconnecting with friends and colleagues at #CCN2025 in Amsterdam and presenting our latest #expertise work.
We 👀 into how #chess experts represent the board, and how the content, structure, and location of these repr shift w/ expertise.⬇️
We 👀 into how #chess experts represent the board, and how the content, structure, and location of these repr shift w/ expertise.⬇️
August 19, 2025 at 9:40 PM
It was great reconnecting with friends and colleagues at #CCN2025 in Amsterdam and presenting our latest #expertise work.
We 👀 into how #chess experts represent the board, and how the content, structure, and location of these repr shift w/ expertise.⬇️
We 👀 into how #chess experts represent the board, and how the content, structure, and location of these repr shift w/ expertise.⬇️
Reposted by Andrea Costantino
Neural manifold properties can help us understand how animal brains deal with competing and multifaceted information, execute flexible behaviors and reuse common computations, writes @mattperich.bsky.social.
#neuroskyence
www.thetransmitter.org/neural-dynam...
#neuroskyence
www.thetransmitter.org/neural-dynam...
Neural population-based approaches have opened new windows into neural computations and behavior
Neural manifold properties can help us understand how animal brains deal with complex information, execute flexible behaviors and reuse common computations.
www.thetransmitter.org
August 4, 2025 at 2:56 PM
Neural manifold properties can help us understand how animal brains deal with competing and multifaceted information, execute flexible behaviors and reuse common computations, writes @mattperich.bsky.social.
#neuroskyence
www.thetransmitter.org/neural-dynam...
#neuroskyence
www.thetransmitter.org/neural-dynam...