Pinned Tweet
New preprint! 🤖🧠
The cost of thinking is similar between large reasoning models and humans
👉 osf.io/preprints/psya…
w/ Ferdinando D'Elia, @AndrewLampinen, and @ev_fedorenko (1/6)
The cost of thinking is similar between large reasoning models and humans
👉 osf.io/preprints/psya…
w/ Ferdinando D'Elia, @AndrewLampinen, and @ev_fedorenko (1/6)
4
26
94
15,496
Andrea de Varda retweeted
Causal reasoning lies at the core of human intelligence but few studies have tested its neural basis. In a new study, we find that despite its importance, causal reasoning is implemented not in a single unified brain system, but rather in separate domain-specific regions.
1
6
20
1,253
Andrea de Varda retweeted
Had a wonderful time giving a guest lecture at Caltech on the cognitive architecture of LLMs, and more broadly, how we can use AI models as model organisms to study intelligence! :)
Many thanks to Prof. @CFCamerer for the invitation!
We discussed the emergent modularity in LLMs, what the similarities between human brains and LLMs might tell us about general principles of intelligence, and how AI models could help us study cognition.
Really enjoyed the discussions and questions! :)
Slides: docs.google.com/presentation/d…
(And special thanks to @devarda_a for letting me steal some of his beautiful slides! :))
Many thanks to Prof. @CFCamerer for the invitation!
We discussed the emergent modularity in LLMs, what the similarities between human brains and LLMs might tell us about general principles of intelligence, and how AI models could help us study cognition.
Really enjoyed the discussions and questions! :)
Slides: docs.google.com/presentation/d…
(And special thanks to @devarda_a for letting me steal some of his beautiful slides! :))
The human brain is strikingly modular, with distinct networks for language, formal reasoning, social reasoning, and physical reasoning. Is this a fundamental principle of how intelligent systems are built, or an accident of biological evolution?
In our latest preprint, we find organization emerges in Large Language Models, another class of intelligent system.
Brains and LLMs are shaped by entirely different kinds of optimization (biological evolution vs. gradient descent). That they arrive at the same modular design anyway suggests modularity may be a fundamental property of intelligent systems.
🌐 Web: pengrui-han.github.io/LLM_Modularity…
📄 Paper: pengrui-han.github.io/LLM_Modularity…
💻 Code & data: github.com/Pengrui-Han/LL…
Using circuit analyses across 46 tasks spanning four cognitive domains, we find:
1️⃣ Tasks that draw on the same network in humans recruit overlapping units in LLMs, while tasks drawing on different networks recruit distinct units.
2️⃣ These units are causally linked to model behavior. Ablating the units critical for one domain impairs performance in that domain (−26% accuracy) but barely touches the others (−2.5%).
This project has been in the works for a while :) Huge thanks to my advisors @jacobandreas @ev_fedorenko @devarda_a, and to @Nancy_Kanwisher for valuable conceptual input and feedback throughout. #MIT
4
19
146
5,836
Andrea de Varda retweeted
If you've noticed a distinct dialect emerging in frontier AI models and struggled to understand what they're talking about, you're not crazy!
We turned to psycholinguistics to investigate what makes the writing style so weird, and why this might be happening
🚨New preprint! 🧵
We turned to psycholinguistics to investigate what makes the writing style so weird, and why this might be happening
🚨New preprint! 🧵
28
136
898
92,674
Andrea de Varda retweeted
New blog post! I wrote about why I think it's useful to use metaphorical (and anthropomorphic) terms to describe AI. I start by describing the role of metaphor in language and cognition, and how metaphors can help us to efficiently understand relations and make predictions 1/2
8
13
161
11,000
Andrea de Varda retweeted
Congrats to @devarda_a, named one of the 2026 #STATWunderkinds by @statnews!
Andrea is doing groundbreaking work in @ev_fedorenko + @roger_p_levy labs to understand how brains and machines process language.
Learn more: statnews.com/wunderkinds/an…
@mitbrainandcog @ScienceMIT
Andrea is doing groundbreaking work in @ev_fedorenko + @roger_p_levy labs to understand how brains and machines process language.
Learn more: statnews.com/wunderkinds/an…
@mitbrainandcog @ScienceMIT
0
3
11
971
Thrilled to be one of @statnews's 2026 #STATWunderkinds for my work on language processing and reasoning in 🧠 and 🤖
Big thanks to my current/previous advisors @ev_fedorenko @roger_p_levy Marco Marelli and everyone I get to work with!
Big thanks to my current/previous advisors @ev_fedorenko @roger_p_levy Marco Marelli and everyone I get to work with!
10
6
64
3,281
Andrea de Varda retweeted
Why do neural networks become modular? Our paper is now out in Nature Machine Intelligence! @NaturePortfolio We show that multi-task learning promotes modularity under limited capacity and task demands help explain features of brain architecture in addition to the spatial anks to all the co-authors. @WHYWYH007 @yifannnwu @DaniSBassett @marcelomattar
rdcu.be/YJ12NBxvrHsa
rdcu.be/YJ12NBxvrHsa
8
56
247
11,174
Check out CogGym, led by @LanceYing42! 258 experiments from 100 cogsci papers (!!) in one format. Main takeaway: models are getting more human-like, but way slower than improvements on math and coding
Have we built machines that think like humans? 🧠🤖
In a new preprint, we propose CogGym, which compares AI and human judgments across 258 commonsense-reasoning experiments from 100 cognitive-science papers.
Paper: arxiv.org/abs/2609.21259
Platform: coggym.org
0
4
23
1,905
Andrea de Varda retweeted
📄👀 Check out our new survey on Eye Tracking and NLP, where we reflect on past work in this space, and outline a roadmap for what's next.
Not needless to say - it's fully human written. The humans are Lena Jäger, David Reich, Omer Shubi and me. zenodo.org/records/229331…
Not needless to say - it's fully human written. The humans are Lena Jäger, David Reich, Omer Shubi and me. zenodo.org/records/229331…
0
2
30
1,877
Andrea de Varda retweeted
📢 The CORE COGNITION lab (core-cognition.github.io) is looking to hire another PhD student, working on automated scientific discovery and knowledge representation in cognitive science.
Starting date: March to September 2027.
🙏 Please share widely!
Starting date: March to September 2027.
🙏 Please share widely!
3
40
164
10,860
New preprint! 🤖➕➗
Shared circuits predict whether LLMs generalize across formats in arithmetic reasoning
w/ @sanapandey, @pengrui_han, @jacobandreas, and @ev_fedorenko
If you can solve 2+5, you can solve "two plus five" without learning math again. LLMs are more brittle (1/5)
Shared circuits predict whether LLMs generalize across formats in arithmetic reasoning
w/ @sanapandey, @pengrui_han, @jacobandreas, and @ev_fedorenko
If you can solve 2+5, you can solve "two plus five" without learning math again. LLMs are more brittle (1/5)
1
22
160
7,502
Circuit overlap predicts generalization at three levels: which formats are hardest (English > Spanish > Italian), which models generalize best, and which individual items get solved (fig below). If the numeric circuit is not engaged, the answer is almost always wrong. (4/5)
1
0
6
477
The circuit metric (unsupervised) predicts generalization as well as or better than supervised probes.
Behavioral generalization is grounded in circuit generalization: if the circuit does not transfer the ability does not either.
Pre-print 🔗 arxiv.org/abs/2609.04463 (5/5)
Behavioral generalization is grounded in circuit generalization: if the circuit does not transfer the ability does not either.
Pre-print 🔗 arxiv.org/abs/2609.04463 (5/5)
1
0
7
413
Andrea de Varda retweeted
Go, @devarda_a! A beautiful and comprehensive study!🔑 finding: behav. measures are ~fully reducible to simple predictors of effort (surprisal, word length+frequency), but for 🧠 measures, LLM embeddings carry additional predictive power, likely capturing aspects of meaning.
New preprint! 🧠👀🤖
Behavioral and brain responses to language reflect different levels of linguistic representation
w/ @whylikethis_ , @ev_fedorenko , and @roger_p_levy (1/10)
1
9
52
3,710
Andrea de Varda retweeted
Wonderful work by @devarda_a linking behavioral and neural manifestations of language comprehension using language models. With @roger_p_levy and @ev_fedorenko
New preprint! 🧠👀🤖
Behavioral and brain responses to language reflect different levels of linguistic representation
w/ @whylikethis_ , @ev_fedorenko , and @roger_p_levy (1/10)
0
2
13
833
A similar asymmetry emerges within the effort measures. Eye movements are driven mostly by length and frequency (context-independent). Brain responses are driven mostly by surprisal (context-dependent). (9/10)
1
0
2
248
👀!=🧠
A unified theory needs both kinds of data with a clear understanding of which levels of representation each measure reflects. (10/10)
Pre-print 🔗 tinyurl.com/mr3cre4b
A unified theory needs both kinds of data with a clear understanding of which levels of representation each measure reflects. (10/10)
Pre-print 🔗 tinyurl.com/mr3cre4b
0
0
5
346
New preprint! 🧠👀🤖
Behavioral and brain responses to language reflect different levels of linguistic representation
w/ @whylikethis_ , @ev_fedorenko , and @roger_p_levy (1/10)
Behavioral and brain responses to language reflect different levels of linguistic representation
w/ @whylikethis_ , @ev_fedorenko , and @roger_p_levy (1/10)
3
16
75
7,459
👀 Behavior: effort dominates. SFL (three numbers) performs better than 1600-dim embeddings, esp. in eye-tracking. Adding EMB to SFL gives only small gains.
🧠 Brain: the opposite. Embeddings predict fMRI/N400 responses better than effort.
(8/10)
🧠 Brain: the opposite. Embeddings predict fMRI/N400 responses better than effort.
(8/10)
1
0
5
349
