I'm again grateful to @theverge.com for publishing my newest long-form essay on the limits of the computational model of the mind, the benefits of understanding both biological and cultural evolution to explain human thinking, and the danger of cognitive hot dogs in education.
Our minds aren’t equipped to handle AI
AI is junk food for the mind: easy, tempting and ultimately very bad for you.
theverge.com
We've just released a new version of childes-db, my lab's interface to the CHILDES database of child language transcripts. It lets you work with CHILDES data from R through a versioned, reproducible API. A few updates 🧵 childes-db.stanford.edu/
It's "advice to grad students" season! Here's a post I wrote several geological epochs ago, in 2019. My advice gets harder and harder to follow every year (even for me). But I think that means it gets better and better? kensycooperrider.com/blog/advice-...
Advice to a young scholar — Kensy Cooperrider
Ten tips for young scholars on doing good work (and living well at the same time).
kensycooperrider.com
Understanding how learners conclude “X laughed Y” is incorrect is an age-old question, with several hypotheses, some of which have been ~impossible to disentangle! @tomyxw.bsky.social, @fredashi.bsky.social, and I use controlled rearing to shed light on this in our new EMNLP paper: 1/n
New paper out in @openmindjournal.bsky.social! "Large Language Models as Distributional Baselines for Language Tasks". With @jamichaelov.bsky.social , @camrobjones.bsky.social , Tyler Chang, and Ben Bergen. direct.mit.edu/opmi/article...
Large Language Models as Distributional Baselines for Language Tasks
Abstract. In order to ask questions about the mechanisms underpinning human cognition, researchers must control for properties of stimuli that could confound detected effects. In experiments involving...
direct.mit.edu
Yes, exactly. Platonic representations are one of the many erroneous conclusions reached under the assumption that intelligence is a one-dimensional scalar.
Victoria Bosch@initself.bsky.social · last mo.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
Almost no psychology research is based on a true probability sample. The question is whether that undermines your research! Ch 10 of Experimentology argues that it depends on whether the effect you're studying is heterogeneous. 🧵 experimentology.io
Great chapter on a very important topic. Construct validity continues to be one of the biggest challenges in research characterizing the behavior of LLMs as well.
Mike Frank@mcxfrank.bsky.social · 3mo agoIf you don't know how reliable your measure is, you're wasting your participants' time. Ch 8 of Experimentology argues that measurement reliability and validity deserve more attention than most experimentalists give them. 🧵 experimentology.io
Will be presenting this work measuring behavioral sensitivity to false belief manipulations in a sample of open-weight LMs at tomorrow's poster session (11am). Stop by if you're interested! #ACL2026 aclanthology.org/2026.acl-lon...
Most of the statistical tests you learned in stats class — t-test, ANOVA, correlation — are actually special cases of a single thing: linear regression. Ch 7 of Experimentology argues that thinking in models, not tests, is more flexible and a better foundation for theory. 🧵 experimentology.io
Seems like a good time to share our new preprint about model openness! (with @catherinearnett.bsky.social @tylerachang.bsky.social Pamela D. Rivière, Samuel M. Taylor @camrobjones.bsky.social @seantrott.bsky.social @rplevy.bsky.social Ben Bergen, and Micah Altman): arxiv.org/abs/2603.26539
In film, "we'll fix it in post" is what you say when something went wrong on set and you don't want to redo it. AI research has made it our entire methodology: train the model, then patch whatever comes out. Our new ICML oral argues this can't be the basis of a science of AI. 🧵
If you're a faculty, research scientist, postdoc, or senior PhD in any area of science, select your top 6 virtues at this link: shorturl.at/bV2lF You can also tell us if you think we missed any. We want broad participation, so pls RT! 🙏 In collab w/@devezer.bsky.social @statmodeling.bsky.social
Epistemic Virtues for Scientific Inquiry — Community Survey
In Six Memos for the Next Millenium, Italo Calvino proposed six literary virtues he felt should be preserved and cultivated in literature regardless of how the world changed: lightness, quickness, exa...
shorturl.at
Very excited to announce that I'll be starting as an Assistant Professor in the Psychology department at Rutgers University-Newark in January 2026!
Will be presenting a new paper on generalizability in mechinterp research at the 2025 NeurIPS MechInterp workshop! Thread below. #NeurIPS
Does vision training change how language is represented and used in meaningful ways?🤔The answer is a nuanced yes! Comparing VLM-LM minimal pairs, we find that while the taxonomic organization of the lexicon is similar, VLMs are better at _deploying_ this knowledge. [1/9]
I think understanding which factors lead to convergence and divergence (both in behavior and internal mechanisms) across networks is crucial to understanding what kinds of systems we're studying and what kinds of claims we can generalize across model instances. Very cool work!
Ann Huang@annhuang42.bsky.social · 11mo ago📍Excited to share that our paper was selected as a Spotlight at #NeurIPS2025! arxiv.org/pdf/2410.03972 It started from a question I kept running into: When do RNNs trained on the same task converge/diverge in their solutions? 🧵⬇️
📍Excited to share that our paper was selected as a Spotlight at #NeurIPS2025! arxiv.org/pdf/2410.03972 It started from a question I kept running into: When do RNNs trained on the same task converge/diverge in their solutions? 🧵⬇️
A confounding thing for the linguistics of LMs: the best way to assess their grammatical ability is string probability. Yet string probability and grammaticality are famously not the same! Really excited to have this out, where we give a formal account, w/ experiments, of how to make sense of that!
Jennifer Hu@jennhu.bsky.social · 11mo agoNew work to appear @ TACL! Language models (LMs) are remarkably good at generating novel well-formed sentences, leading to claims that they have mastered grammar. Yet they often assign higher probability to ungrammatical strings than to grammatical strings. How can both things be true? 🧵👇
Hard to process the news about Harvard and international students. Other universities should stand in solidarity with our colleagues who are being persecuted.
Please read my essay in TIME, which @science.org did not do carefully before publishing this assertion. time.com/7285045/resi...
Why I’m Resigning from the NSF and Library of Congress
I cannot participate in systems that require dishonesty as the price of belonging.
time.com
My awesome colleague @seantrott.bsky.social quoted on understanding the nature of our new robot overlords in @nytimes.com www.nytimes.com/2024/12/16/o...
Opinion | How Does A.I. Think? Here’s One Theory.
It may employ a kind of reasoning humans use all the time — but you’ve probably never heard of it.
nytimes.com