Postdoc at MIT BCS, interested in language(s) and thought in humans and LMs
Joined March 2022
- Tweets 158
- Following 529
- Followers 445
- Likes 1,278
24 Photos and videos
Pinned Tweet
25 Jul 2025
New preprint! 🤖🧠
The cost of thinking is similar between large reasoning models and humans
👉 osf.io/preprints/psyarxiv/m2…
w/ Ferdinando D'Elia, @AndrewLampinen, and @ev_fedorenko (1/6)
4
26
93
14,578
Andrea de Varda retweeted
It is tricky to characterize the features represented by human language cortex. This work is a step toward doing so.
Using small, interpretable feature sets, we explain language-network responses and show a shared feature basis across regions with variation across individuals.
🚨New preprint!🚨
We know that LM representations can be used to predict brain responses to language. But what *features* of these representations underlie this alignment? We use SAEs to find out!
10
64
6,061
Andrea de Varda retweeted
Jun 9
Excited to announce that I'll be starting my own lab in Tübingen this October!
Hiring at all levels: Postdoc, PhD & RA.
Want to work on computational cognitive science at scale? Apply: core-cognition.github.io/
Reposts and shares much appreciated 🙏
16
47
333
27,071
Andrea de Varda retweeted
Congrats to Andrea de Varda for winning the 2026 Glushko Dissertation Prize! This highly competitive award recognizes outstanding young researchers in the field of cognitive science. Andrea studies how language is represented in human brains and machines. cognitivesciencesociety.org/…
ALT Photo of smiling researcher with clasped hands next to words that read: ANDREA DE VARDA WINS GLUSHKO DISSERTATION PRIZE- OUTSTANDING DISSERTATION IN COGNITIVE SCIENCE
1
7
532
Thrilled that my thesis on multilingual LLMs in cog sci won a Glushko prize! Huge thanks to my advisor Marco Marelli who supported me throughout and to @ev_fedorenko who I did the multilingual brain work with. Thanks also to @cogsci_soc and the Glushko-Samuelson Foundation! 🙏
Congratulations to Yang ICoN fellow Andrea de Varda @devarda_a, who has been named a winner of the prestigious Glushko Dissertation Prize! 👏👏
🔗 cognitivesciencesociety.org/…
@cogsci_soc
1
5
41
5,930
While I'm here: I think multilingual LLMs are vastly underrated as scientific tools for studying language. They have to represent hundreds of languages in one set of weights, so whatever those languages share gets compressed into the same place. I think that is very cool!
4
219
Andrea de Varda retweeted
We don't start out with a bilateral language system: it's already strongly left-lateralized in young kids! Resilience of language to early LH damage must occur in spite of this early hemispheric bias. Congrats to Ola @olaozpa and Amanda @Amanda_M_OBrien! Out in NatComms now! 🎉
May 18
Excited to share that our paper: ‘Precision fMRI reveals that the language network exhibits adult-like left-hemispheric lateralization by 4 years of age’ is finally published!
mcgovern.mit.edu/2026/05/17/…
nature.com/articles/s41467-0…
@Amanda_M_OBrien @ev_fedorenko
12
47
5,608
Andrea de Varda retweeted
Apr 21
🎉 Happy to share that our paper on function words & language learning (w/ Heidi Getz & @weGotlieb) is accepted to #ACL2026! A little late to the party, but still worth celebrating 🥳
We ask: what statistical properties help a learner abstract grammatical knowledge from linear input? Turns out function words, though often overlooked, play an important role.
Check out our updated preprint: arxiv.org/pdf/2601.21191
🧵 1/4
2
14
73
5,009
Andrea de Varda retweeted
TL;DR: Languages look the way they do because our brains find them easier to handle that way. 🧠✨
Mar 16
Why do languages share common properties? Adults learned novel quantifiers satisfying semantic universals faster than those violating them. This suggests that learnability helps explain why certain meanings are lexicalized across cultures.
@Logic_Cognition
9
36
2,945
Andrea de Varda retweeted
Mar 17
Language, Intelligence & Thought lab is looking for a lab manager! This is a 2-year postbac position that will allow you to gain experience in human neuroscience, cognitive science, and AI research prior to applying to PhD programs.
Express interest here: forms.gle/289sLgZdJ2bQr1Y48
1
27
58
9,095
Andrea de Varda retweeted
Can we process meaning unconsciously? Our new study suggests: not really… unless language has a way to express it!
New paper out with Andrea Nadalini @D_Casasanto @CrepaldiDavide @BottiniRob
1
2
7
1,112
Andrea de Varda retweeted
Mar 10
📢 PhD position in Developmental Language Modelling (plz RT🙏)
What can human language acquisition teach us about training language models? Join us as a PhD!
4 yrs, fully funded, MPI-NL; april 3
mpi.nl/career-education/vaca…
3
28
105
19,677
Andrea de Varda retweeted
Short post on what I call the "no-magic approach to understanding intelligent systems" — the philosophy I think of as motivating our work on understanding intelligence without resorting to magical thinking about AI or humans! Link below:
ALT Title: The no-magic approach to understanding intelligent systems Image: an abstract watercolor of a neural network at the bottom of a loss landscape
8
57
388
27,665
Andrea de Varda retweeted
Mar 5
📢 PhD position in the NeuroAI of Language
Why can LLMs predict brain activity so well? We're hiring a PhD student to find out -- AI interpretability meets neuroimaging
Deadline March 20.
Please RT 🙏
mpi.nl/career-education/vaca…
5
26
180
21,023
Andrea de Varda retweeted
Feb 4
My PhD thesis is out 🥳🎓
How do LLMs, trained on trillions of tokens, reason?
Can they generalise beyond their training data or are they constrained by what they've seen before?
My takeaway: they can generalise beyond training in interesting ways, showing genuine reasoning
94
237
1,986
106,385
Andrea de Varda retweeted
Our researchers don't just study the brain - they help young students see themselves as future neuroscientists. @mitbrainandcog research scholar Zadriana Smith postdoc @HalieOlson recently took time away from their labs to inspire the next generation of neuroscientists! ✨
ALT BCS research scholar Zadriana Smith shows a group of middle schoolers what an actual research brain looks like.
ALT Postdoc Halie Olson holds a tiny model brain to illustrate to a group of middle schoolers what makes kid brains different than adult brains.
4
12
1,971
Andrea de Varda retweeted
New in @NeuroCellPress! A team including #KempnerInstitute’s @_coltoncasto & @GretaTuckute maps the cerebellum's role beyond motor control as part of an extended language network.🧠🗣️
More here: bit.ly/4rptQ13 #neuroscience #fMRI @harvardmed @HarvardGSAS @ev_fedorenko
3
11
3,283
Andrea de Varda retweeted
Fed up with hot anecdotal takes on AGI? 🤖🧪
Join us in Warsaw for actual science on how models reason.
📅 Jan 8, Warsaw🔗 Program: manuel-vg.github.io/composit…
Featuring @DieuwkeHupkes @MilekPl @leo_bertolazzi @IPrattHartmann @devarda_a and others.
3
6
552
Andrea de Varda retweeted
24 Dec 2025
LLMs develop novel biases from experience.
New preprint: LLMs that make decisions & get feedback develop new views — including ⚠️harmful stereotypes that target demographics!
[1/7]
1
3
27
13,252
Andrea de Varda retweeted
I may be a *little* biased but this 📘 is GREAT! If you ever found language structure interesting, but were turned off by implausible overly complicated accounts, this book is for you: a simple and empirically grounded account of the syntax of natural languages. A must-read!
24 Dec 2025
New book! I have written a book, called Syntax: A cognitive approach, published by MIT Press.
This is open access; MIT Press will post a link soon, but until then, the book is available on my website:
tedlab.mit.edu/tedlab_websit…
2
15
68
10,274
25 Dec 2025
It's fascinating that you can explain *so much* with dependency distance (effects in language production, comprehension, cross-linguistic differences in word orders, the difficulty of 'legalese'...). Highly recommended!
24 Dec 2025
New book! I have written a book, called Syntax: A cognitive approach, published by MIT Press.
This is open access; MIT Press will post a link soon, but until then, the book is available on my website:
tedlab.mit.edu/tedlab_websit…
3
9
3,163