Sam Gershman

Unfortunately it’s clear that the Israeli people can’t topple the government or change its ways. Seems like the attorney general, the last gate keeper, will be fired soon. As long as we’re stuck with this government, the war won’t stop. Only international intervention (=the U.S) might do it.

Entorhinal cortex signals dimensions of past experience that can be generalised in a novel environment https://www.biorxiv.org/content/10.1101/2025.08.01.668096v1

Wow. A group of top scholars at Harvard just sent a letter to its president, Alan Garber, warning against surrendering to Trump.

Signatories include Steven Levitsky, Dani Rodrik, Ryan Enos, Theda Skocpol, and Steven Walt.

Someone forwarded it to me. Read it here:

A wonderful article on work (both old and new) pushing the frontier of memory research.

Seen at @nikolaskaralis.bsky.social

#neuroskyence #psychSciSky #sensorimotor

Excited to be sharing my latest work with @tobigerstenberg.bsky.social at #CogSci2025!

Learning usually occurs when we encounter new data. But we also have the capacity to reflect on our past experiences. What can we learn from simulating past experience?

📃 cicl.stanford.edu/papers/yang2...

1/

Journal editor receiving a submission right before their holiday.

This is unfortunate:
www.nytimes.com/2025/07/28/u...
I wrote to the president of Harvard. I hope other faculty will speak their conscience, even if it means more struggle ahead.

Looking forward to sharing our work at #cogsci2025! We aim at getting one step closer to a domain-general formulation of mental costs via policy compression.

Come see my presentation at "Talks 35: Reasoning". It is scheduled at 16:44 PST, August 1 at Nob Hill C!

gershmanlab.com/pubs/LiuGers...

I remember having dinner with her (and a group of people) after a conference and she exclaimed astonishment that anyone would ever cite a book or paper without reading it fully (Mind as Machine had thousands of references). Now I feel a pang of guilt every time I do that.

Her book Mind as Machine was magnificent, the best history of cognitive science by a mile. It also provoked an amusing argument with Chomsky (his reviewing style didn't work quite as well on Boden as it did on Skinner).

🚨📊ɴᴇᴡ ᴘᴀᴘᴇʀ!! When we learn a new skill—like tying shoelaces or making pasta—we often start with a series of deliberate steps. But eventually, those steps blend into a smooth, single unit: an action chunk. But why and when do we chunk? 👇
📰https://authors.elsevier.com/a/1lMMD2Hx2-9B8

Feature request for FaceTime: 'grandma mode' where it crops everything but your forehead.

Our latest on the cognitive science of LLMs! To be presented @CogSci‬2025 🎉

LLMs are increasingly involved in human collaborations. How do LLMs assign responsibility and reward to collaborators? Is it similar to how humans do it? 🤖🧑

📃 gershmanlab.com/pubs/XiangBi... (1/4)

Sorry, I'm not quite following this. I read your second post first and was trying to figure out the point you're trying to make. Maybe a visual aid would help?

I wonder if I could make a class with no syllabus open only to undergrads planning to go into finance/consulting after graduation. It might also be good for pre-med and pre-law students.

📣 I'm looking for a postdoc to join my lab at NYU! Come work with me on a principled, theory-driven approach to studying language, learning, and reasoning, in humans and AI agents.
Apply here: apply.interfolio.com/170656
And come chat with me at #CogSci2025 if interested!

GILLETTE: now with 5 blades for a smoother closer shave

OCCAM: what the fuck

I've been wanting for years to teach a "class with no syllabus" as an exercise in intellectual spontaneity. Maybe I will. But I fear that it won't really move the needle when all the social forces are pushing in the other direction.

I’m a Student at Harvard. Almost Everyone I Know Wants the Exact Same Type of Job. Contrary to popular belief, Ivy League students are not overly idealistic. Quite the opposite.

Not surprising, but still depressing:
slate.com/life/2025/07...
What's fascinating to me is the cultural transmission of risk aversion. I doubt all these students entered Harvard believing their destiny was finance/consulting.

So excited our paper is now out in ‪@cognitionjournal.bsky.social‬! Huge thanks to our editor and reviewers 🧠 Their thoughtful suggestions inspired Experiments 3 & 4, including a striking inverse correlation between idleness judgments and speed-up predictions

Fast efficient coding and sensory adaptation in gain-adaptive recurrent networks As the statistics of sensory environments often change, neural sensory systems must adapt to maintain useful representations. Efficient coding prescribes that neuronal tuning curves should be optimize...

Efficient coding theories often implicitly or explicitly assume slow changes in tuning (e.g., through synaptic plasticity). Arthur Prat-Carrabin has collected psychophysical data showing that it can be fast, and this can be explained by a gain-adaptive RNN:
www.biorxiv.org/content/10.1...

By the way, it would be interesting to connect this work to earlier work by Eric Bigelow and @spiantado.bsky.social using data from people playing the number game:
escholarship.org/content/qt43...

Very cool!

"replacements that are aligned with the priorities of the administration"?? Who are these replacements... scientists? Any scientist who would bend their judgment to political priorities isn't a scientist at all.

Apparently in the 1970s Richard Solomon (progenitor of opponent-process theory and member of the National Academy of Sciences) was allowed to wander around a hospital nursery making random newborns cry.

(source: people.whitman.edu/~herbrawt/cl...)

Thinking about writing a pop science book about the brain (“Your Brain on Everything”) where the big reveal is that all the brain names were randomized and I was actually talking about psychology the whole time.

Blake, didn't you argue in that paper that maybe the soma encodes prediction errors? Are you arguing now against the idea that there are prediction errors at all in sensory cortex, or just against the idea there are dedicated error neurons?