Sign in

Vlad Ayzenberg

@vayzenb.bsky.social
3.2K followers 685 following 167 posts

Director of the Vision Learning and Development Lab at Temple University. Interested in cognition, computation, neuroscience, and development. vlad-lab.com

PostsRepliesMedia
Vlad Ayzenberg @vayzenb.bsky.social · 19h
Original thread here: bsky.app/profile/vayz...
010
Vlad Ayzenberg @vayzenb.bsky.social · 19h
Excited to have been part of this great special issue. Our review described the bidirectional impacts of developmental science on AI www.cell.com/neuron/fullt...
cell.com
Advancing NeuroAI through developmental alignment
Ayzenberg et al. describe how principles from child development can be used to improve the capacities and biological plausibility of AI models. These principles can be incorporated into every stage of...
150
Reposted by Vlad Ayzenberg
MindCORE at the University of Pennsylvania @pennmindcore.bsky.social · 08/10/2026
The MindCORE Fellows Application is open! We seek applications from recent PhDs in neuro, psych, ling, philo, comp sci, or other cog sci discipline for our 3-yr postdoc. MindCORE is Penn's interdisciplinary effort to understand human intelligence & behavior. mindcore.sas.upenn.edu/post-doctora...
11914
Reposted by Vlad Ayzenberg
Deon T. Benton @deontbenton.bsky.social · 07/10/2026
🚨 New publication at Nature Reviews Psychology! (natrevpsychol.nature.com) rdcu.be/d2jxg7tYHWNa A recently published paper in the same journal argued that infants use two separate domain-specific systems -- naive psychology and naive physics -- to reason about objects and others' minds.
rdcu.be
1106
Reposted by Vlad Ayzenberg
sarah tung @sarahtung.bsky.social · 05/10/2026
1/ New preprint! 🧠👶 "Eccentricity, not category, organizes infant ventral temporal cortex." With Emily M Chen, Cleo Tay, Kylie Syzmanski, and @camerontellis.bsky.social www.biorxiv.org/content/10.6...
biorxiv.org
Eccentricity, but not category, organizes infant ventral temporal cortex
Scene- and word-selective regions in adult ventral temporal cortex occupy consistent locations across individuals and align with eccentricity biases for peripheral versus central parts of the visual f...
15216
Reposted by Vlad Ayzenberg
Andrey Chetverikov @achetverikov.bsky.social · 02/10/2026
Some months ago, I wrote a skeptical reply to a commentary on bots in online studies in PNAS, arguing that no evidence yet exists for such bots. Together with the authors of the original commentary, we now show that bots can pass (almost?) all online cognitive studies with human-like performance 1/n
25129
Reposted by Vlad Ayzenberg
Vaughn Cooper @vscooper.micropopbio.org · 25/09/2026
FYI to everyone submitting this October NIH cycle You need to re-certify your ScienCV AFTER Sept 26 (tomorrow) for your application or it may not be accepted. Great timing!
78592
Reposted by Vlad Ayzenberg
Michelle Greene @mgreenephd.bsky.social · 25/09/2026
🚨New paper alert! 🚨 The visual system tunes itself to the statistics of its input. How do individual experiences differently tune vision? In this work, @bjbalas.bsky.social and I examined how one individual difference—height—changes the visual diet and how this affects perception. 1/
29432
Reposted by Vlad Ayzenberg
Tom Olino @olinotom.bsky.social · 23/09/2026
I will be reviewing applications this admissions cycle (pending funding). This time, I'll be looking for folks interested in research within our training clinic. We have implemented ROM, and want to build on that infrastructure. Other research is possible; the primary goal is clinical research.
1119
Reposted by Vlad Ayzenberg
Grace Lindsay @neurograce.bsky.social · 27/08/2026
MIT has put out an "AI and Education" report summarizing what they've found at their own institution. Seems thoughtful. aiandeducation.mit.edu/report/
421968
Reposted by Vlad Ayzenberg
Jake Quilty-Dunn @quiltydunn.bsky.social · 11/09/2026
New paper, co-authored with Justin Wood, in @cp-trendscognsci.bsky.social: The Nativist-Empiricist Debate Is Broken We argue that radical nativism and radical empiricism are compatible and, in some cases, both plausibly true. So something is wrong. 🧵 Author share link here, valid til Oct 31:
authors.elsevier.com
Please wait whilst we redirect you
All content on this site: Copyright © 2026 Elsevier B.V., its licensors, and contributors. All rights are reserved, including those for text and data mining, AI training, and similar technologies. For all open access content, the relevant licensing terms apply.
17630
Reposted by Vlad Ayzenberg
Rhodri Cusack @rhodricusack.bsky.social · 10/09/2026
We're hiring! 2 postdocs + 3 PhD students to join InfantNeuroAI at Trinity College Dublin: awake infant fMRI & OPM-MEG, plus computational models of how infants learn. Aims: understand infant visual cognition, infant neuroimaging methods, more energy-efficient learning 🧵 www.cusacklab.org/vacancies
16652
Reposted by Vlad Ayzenberg
Marco Barilari @marcobarilari.bsky.social · 10/09/2026
New preprint from the CPPlab - our first 7T MRI study How does sound reach the occipital cortex in early blind people? Using 7T VASO layer-fMRI, here the first functional sub-mm dataset in blind people to test if auditory motion engages hMT+/V5 layers via feedforward or feedback lnk.ua/kjYccwS9Y 🧵👇
13213
Reposted by Vlad Ayzenberg
Jenny Wang 👶 🧮 @jinjingjenny1.bsky.social · 10/09/2026
seeking help 🆘 a senior colleague advised that i need to give more invited talks to strengthen my tenure case... how do i find such opportunities?
static.klipy.com
Wirt from Over The Garden Wall Looks Awkward
ALT: Wirt from Over The Garden Wall Looks Awkward
396
Reposted by Vlad Ayzenberg
Daniel Lakens @lakens.bsky.social · 03/09/2026
A good time to remind people that 1) not all deviations from a preregistration are equally problematic for the severity of a test, and 2) if you need to deviate, deviate. The validity of the test trumps its severity. online.ucpress.edu/collabra/art...
0206
Reposted by Vlad Ayzenberg
Biyu Jade He @biyuhe.bsky.social · 01/09/2026
📣 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).
26649
Reposted by Vlad Ayzenberg
Victoria Bosch @initself.bsky.social · 28/08/2026
Are brains and artificial neural networks converging onto universal representations? There is a seductive idea making the rounds in NeuroAI / machine learning: train systems well enough, and they all converge on the same representation of reality (i.e. a unique world model). We have thoughts™ 1/n
cell.com
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 univer...
818175
Reposted by Vlad Ayzenberg
Sami Yousif @samiyousif.bsky.social · 26/08/2026
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...
sciencedirect.com
Cognitive maps as a medium for thought
Cognitive maps are essential for navigation, but to what extent are they involved in thinking more broadly? In this opinion article, we examine whethe…
29528
Reposted by Vlad Ayzenberg
Jorge Morales @jorgemorales.me · 26/08/2026
The NSF GRFP solicitation is out. Aspiring Psychology PhD students, the deadline is October 20. www.nsf.gov/funding/oppo...
nsf.gov
NSF Graduate Research Fellowship Program (GRFP)
01610
Reposted by Vlad Ayzenberg
Stand Up for Science! @standupforscience.net · 24/08/2026
Today is the LAST DAY to leave your public comment on the NSF Proposal. The Proposal implements Vought’s science destroying OMB Policies ahead of time, and we cannot let this sneaky proposal scoot by quietly. Leave your public comment today by visiting the link: zurl.co/ZGP3i
The graphic contains a photo of Russell Vought, the red Stand Up for Science logo, and the NSF logo. The graphic reads: "Today is the last day to leave your public comment on the science endangering NSF Proposal".
497107
Reposted by Vlad Ayzenberg
Lucas Benjamin @lucaswbenjamin.bsky.social · 24/08/2026
Our new paper is now out in @pnas.org! How does the (baby) brain extract regularities from sequences from what-follows-what up to large-scale network structure? What if it all came from a single mechanism? www.pnas.org/doi/abs/10.1...
15216
Reposted by Vlad Ayzenberg
Nora Newcombe @noranewcombe.bsky.social · 21/08/2026
One reason why I worry that "experimental psych" memory tasks are not capturing all of episodic memory development. Free reading link for 50 people. www.tandfonline.com/eprint/FAAGH...
tandfonline.com
The Role of Narrative Structure in Scaffolding Children’s Recall
Children’s episodic memory improves sharply between 4 and 7 years. Here, we asked whether the developing ability to impose narrative structure on events may be one reason for this change. Children ...
0229
Reposted by Vlad Ayzenberg
Stand Up for Science! @standupforscience.net · 19/08/2026
5 DAYS LEFT TO LEAVE YOUR PUBLIC COMMENT ON THE NSF PROPOSAL. This is a sneaky, early implimentation of Vought’s OMB Proposal policies. Have your voice heard - visit the link to leave your comment today: zurl.co/JgPtq #StandUpForScience #Vought #NSF #Science #SaveScience
The graphic displays a photo of Russell Vought. The graphic reads: "5 DAYS LEFT, to leave your public comment on the NSF Proposal. Removed requirements tied to statutory civil rights obligations; Applies international collaboration restrictions; Implements Vought's takeover of federal grants ahead of the OMB Proposal. Use our tool & guide to leave your comment today! LINK IN BIO".
13534
Reposted by Vlad Ayzenberg
Sami Yousif @samiyousif.bsky.social · 18/08/2026
There's been discussion about Prolific's dynamic pricing model here on bluesky. To learn more about this (and to make loads of cash, apparently!) I decided to become an actual Prolific participant. See thread below, but TLDR: What's happening on Prolific is not what I expected!👇
23313
Reposted by Vlad Ayzenberg
Nora Harhen @noraharhen.bsky.social · 11/08/2026
How do we make good decisions with little experience? Adults can rely on memories of past choices. In our new study in @pnas.org we show kids & teens do something similar–but what they remember turns out to be quite different! www.pnas.org/doi/10.1073/...
pnas.org
Developmental changes in memory structure and precision alter the use of retrieved episodes during decisions for reward | PNAS
Most widely studied option evaluation strategies rely on knowledge accumulated across repeated experiences. But how should options be evaluated in ...
29340
Vlad Ayzenberg @vayzenb.bsky.social · 07/08/2026
Nice paper providing causal evidence that the dorsal stream is involved in global shape perception direct.mit.edu/imag/article...
direct.mit.edu
Causal evidence for dorsal stream involvement in object shape recognition: A tractography-guided ccPAS study
Abstract. Visual experience requires orchestration of intricate brain connections. One of the most influential models of human visual information processing posits two segregated streams and postulate...
031
Reposted by Vlad Ayzenberg
Rocco Chiou @rocco-chiou.bsky.social · 04/08/2026
At long last, I’m excited to share our latest preprint on the cortical topography of mental imagery! Building on the tethering hypothesis of brain evolution, we examined the spatial layout of imagery-driven neural activities to know how visual, auditory, tactile & multisensory imagery is organised.
1305
Vlad Ayzenberg @vayzenb.bsky.social · 03/08/2026
Excited for my lab's first @cogcompneuro.bsky.social ! We are presenting a mix of projects, ranging from infant modeling and neuroimaging to one-shot category learning #CCN2026
A list of talks from the Vlad lab
0161
Reposted by Vlad Ayzenberg
Ruth Rosenholtz talks to strangers @ruthrosenholtz.bsky.social · 02/08/2026
x.com/NVIDIAAI/sta...
x.com
NVIDIA AI (@NVIDIAAI) on X
Humans count the boxes in this image, hidden ones included, with 82.1% accuracy. The best off-the-shelf multimodal model manages 17.7%. Spatial-IQ is a diagnostic benchmark from NVIDIA Research that ...
021
Vlad Ayzenberg @vayzenb.bsky.social · 31/07/2026
No judgment here, we had a bad sleeper and went through the whole painful process.
110
Vlad Ayzenberg @vayzenb.bsky.social · 31/07/2026
Wait, what's the potty training book?
100
Vlad Ayzenberg @vayzenb.bsky.social · 30/07/2026
Shape perception in COWS. Love this work from @stellalourenco.bsky.social and Greg Berns Now we just need a egocentric model of COW vision and the story will be complete www.researchsquare.com/article/rs-1...
researchsquare.com
Global and local shape perception in cattle: Evidence from a novel dishabituation paradigm
Understanding how farmed animals perceive and differentiate objects in their environment has important implications for animal cognition and welfare. Previous studies in agricultural species have larg...
030
Reposted by Vlad Ayzenberg
Vlad Ayzenberg @vayzenb.bsky.social · 22/07/2026
Excited to share our review in @cp-neuron.bsky.social with @lauriebayet.bsky.social and @mickbonner.bsky.social! We describe how implementing principles from child development can advance the mechanistic plausibility and capacities of AI models We packed A LOT into this review, here's a quick 🧵
17030
Vlad Ayzenberg @vayzenb.bsky.social · 22/07/2026
Check it out! authors.elsevier.com/a/1nTlH3BtfH...
030
Vlad Ayzenberg @vayzenb.bsky.social · 22/07/2026
There's so much more in this review and we really tried to provide a big picture primer into the large-scale principles that shape cognitive development We really hope this review will be of practical use for modelers, by providing an important source of inspiration for the next generation of AI
160
Vlad Ayzenberg @vayzenb.bsky.social · 22/07/2026
Finally, the developmental literature provides a rich set of benchmarks for model comparison Evaluating the developmental trajectory of models may offer a new metric for biological alignment, and also improve model efficiency by uncovering the minimal conditions necessary for a capacity to arise
Developmental trajectory of face and language perception. Face and language perception are domains where infants exhibit a mixture of adult-like and infant-unique capacities during the first year. The capacity to detect visual and auditory stimuli is present by the start of the third trimester, and fetuses even exhibit selectivity for voices prior to birth. At birth, infants show some signatures of adult-like processing with preferences for upright face-like configurations and linguistic features of their parents’ native language. Infants also exhibit qualitatively different capacities, such that, prior to 6-months, infants show stronger face and linguistic discrimination abilities than adults. By 12-months, these capacities undergo perceptual narrowing, such that they become specialized for the faces and linguistic features that appear in the infant’s environment.
160
Vlad Ayzenberg @vayzenb.bsky.social · 22/07/2026
Third, the objectives of children are open-ended and best described by a desire to seek novelty within an understandable "Goldilocks zone" This can be modeled using predictive frameworks where models prioritize data that progressively reduces error, i.e., where they make the most progress
Predictive learning for scenes of different complexity. Scenes that are too simple or too complex relative to a child’s current understanding show little change in their error rate (Δ) across time: if the scene is too simple, prediction accuracy is at ceiling; if the scene is too complex, prediction accuracy is at floor. By contrast, scenes of intermediate complexity, which are in an understandable range, exhibit a progressively decreasing prediction error across time. Once Δ error asymptotes, there is no additional information to be gained from the stimulus and the child shifts their attention to a new scene. As the child learns, new scenes that were previously too complex become understandable.
181
Vlad Ayzenberg @vayzenb.bsky.social · 22/07/2026
Second, children learn from many fewer examples than current models However, their experiences, starting in the womb, provide an excellent curriculum for learning. And, indeed, training models on a child-like sequence of experience improves their robustness and efficiency
Visual inputs in the womb and through the first year. Infant visual inputs become progressively more complex as they age. In the womb their primary visual inputs are from structured, but spontaneously generated signals from the retina, known as retinal waves. At birth, their vision is dominated by simple visual scenes, like architectural features or stimuli close to their face. At this stage, infant vision is blurrier and of lower contrast and saturation than adults, but is sufficient for important stimuli, like seeing a mother’s face while being held. By six-months-of-age, infants’ postural and manual motor control has sufficiently matured to allow them to sample more complex visual environments and even manipulate objects.
161
Vlad Ayzenberg @vayzenb.bsky.social · 22/07/2026
First, the brain develops along a sensory-association gradient, with early developing areas providing a foundation for the development of higher level ones This maturation process may prevent catastrophic forgetting, and could be implemented in AI by training model layers hierarchically
Hierarchical maturation of cortex and models. (A) The human brain exhibits a hierarchical gradient of development wherein sensorimotor areas mature early in life (yellow) and association areas mature later in life (purple). (B) This pattern may be implemented into AI models using layer-wise learning, wherein the learning rate is progressively adjusted across layers over the course of training. Such a layer-wise learning approach may similarly lead the weights of early layers to stabilize (i.e., mature) before later ones. Figure 1A adapted from Sydnor et al.57.
1124
Vlad Ayzenberg @vayzenb.bsky.social · 22/07/2026
We also offer practical examples for how developmental principles can be implemented in all stages of the modeling process, as well as recommendations for using child data as benchmarks
140
Vlad Ayzenberg @vayzenb.bsky.social · 22/07/2026
We argue that to build intelligence we need to model the process by which human intelligence arose, namely, childhood We describe how the architecture, experiences, and learning objectives of children are different from adults, but optimally suited to support learning
281
Vlad Ayzenberg @vayzenb.bsky.social · 22/07/2026
The central goal of NeuroAI is to uncover the factors that give rise to human-level intelligence Yet, the field has predominantly sought to achieve this goal by modeling adulthood--the end state of intelligence
160
Vlad Ayzenberg @vayzenb.bsky.social · 22/07/2026
Excited to share our review in @cp-neuron.bsky.social with @lauriebayet.bsky.social and @mickbonner.bsky.social! We describe how implementing principles from child development can advance the mechanistic plausibility and capacities of AI models We packed A LOT into this review, here's a quick 🧵
17030
Reposted by Vlad Ayzenberg
Melissa Kibbe @levelsof.bsky.social · 20/07/2026
New from me and Alan Leslie! We got the band back together to lay out our view on object representations in the infant mind: non-discursive indexes that operate within a language of thought, & which get development of a more complex object concept off the ground www.sciencedirect.com/science/arti...
sciencedirect.com
24717
Reposted by Vlad Ayzenberg
Francesco Poli @francescopoli.bsky.social · 13/07/2026
New Preprint! We gave neural networks infant-like curiosity and let them free to explore. They grew brain-like architecture, followed human synaptic development, and even started to display compositional thought. So we think curiosity doesn't just come from a complex mind... it helps us build one 👇
biorxiv.org
Curiosity shapes brain-like architectures and functions
How does complex cognition emerge from simpler underlying processes? We show that two components are sufficient: infant-like curiosity and brain-like biophysical constraints jointly drive the emergenc...
17628
Reposted by Vlad Ayzenberg
Stand Up for Science! @standupforscience.net · 10/07/2026
🚨 HEADS UP: If you haven't submitted a comment on the OMB proposal yet, there's still time - but it's running out! There are only 3 days left to push back against Russell Vought's attempt to give political appointees unprecedented control over federal research funding. Take Action: zurl.co/J92d9
Graphic from Stand Up for Science urging action on a federal funding proposal. Large text reads: "3 Days Left" and "Help Us Stop Vought From Destroying American Science." A portrait of Russell Vought appears beside the U.S. Capitol. A banner along the bottom reads: "Submit a comment on the OMB proposal | Link in bio to take action." The Stand Up for Science logo appears in the top left.
48479
Reposted by Vlad Ayzenberg
Jay Van Bavel, PhD @jayvanbavel.bsky.social · 05/07/2026
We should pay peer reviewers Paying reviewers led to faster decisions—an average of 5.5 working days vs 38 days for unpaid reviews And review quality improved (judged by handling editors) Reviewers who delivered a good peer review in 4 days were paid $290 www.nature.com/articles/d41...
115340
Reposted by Vlad Ayzenberg
Deon T. Benton @deontbenton.bsky.social · 30/06/2026
🚨 Officially out at Child Development! A few years ago in Science, Ashley Thomas (ashleyjthomas.bsky.social) and colleagues argued that saliva could be a cue infants use to infer the strength of social relationships. academic.oup.com/chidev/advan...
academic.oup.com
An associative learning account of how saliva becomes a cue for comfort
Abstract. A recent study by Thomas et al. (2022) showed that 8.5- to 10-month-olds use saliva sharing as an indicator of relationship closeness: infants ex
13612
Vlad Ayzenberg @vayzenb.bsky.social · 18/06/2026
I typically use brainiak.org for searchlights. Speed always felt pretty reasonable to me. It just converts the data to a numpy array and then you can just use scikit or whatever from there, which is pretty well optimized
brainiak.org
Brain Imaging Analysis Kit
110
Vlad Ayzenberg @vayzenb.bsky.social · 14/06/2026
www.photopea.com Is an objectively great alternative to both Photoshop and GIMP. It is nearly identical to Photoshop and runs in the browser. I have access to Photoshop and haven't used it in years.
photopea.com
Photopea | Online Photo Editor
Photopea Online Photo Editor lets you edit photos, apply effects, filters, add text, crop or resize pictures. Do Online Photo Editing in your browser for free!
030