Stefano Palminteri (@stepalminteri.bsky.social)

Few weeks ago I discovered this book (and, on the same occasion, a whole field of research...). This monograph summarises decades of research about the effect of reward magnitude in animal behaviour. Spoiler alert: rewards have relative effects.

I have just finished reading “the new behaviourism” by John Staddon. Here are a few thoughts about it. First all the book is clearly strongly opinionated. So if you are looking for a “balanced” view of behaviorism or the cognitive revolution, you should look elsewhere 1/n

🤖🧠🚨 New preprint on ... consciousness in an artificial system, defending the unpopular opinion that cognitive science is behaviourism: the learn more go and see in bluesky! bsky.app/profile/stepal…

EP #4 | @StePalminteri mentions that students shouldn't hesitate reading the old classics/papers in psychology/neuroscience. He also recommends

We are happy to have Prof Stefano Palminteri @StePalminteri from Institut National de la Santé et de la Recherche Médicale & École Normale Supérieure to give our next talk! For more info, check our website. All welcome. #LJDM25 #Riskpreferences

When we started studying LLMs psychology, we faced the problem of making sense of the results, especially concerning closed models. @nicolas__yax developed a very creative idea into a full-fledged to solve this problem. Great work and collaboration with @pyoudeyer @inria

Congratulations @nicolas__yax for the amazing work, and @StePalminteri ! I have to say that when Nicolas pitched us the idea of this project, I was excited about the creativity, but was not convinced it would work: I was completely wrong, it worked close to directly! Beautiful!

🔥Our paper PhyloLM got accepted at ICLR 2025 !🔥 In this work we show how easy it can be to infer relationship between LLMs by constructing trees and to predict their performances and behavior at a very low cost with @StePalminteri and @pyoudeyer ! Here is a brief recap ⬇️

Link to the paper: osf.io/preprints/psya…

🌶️ Just published a spicy paper with @StePalminteri examining how cognitive biases fundamentally shape the heated discourse around Large Language Models. Our research reveals how cognitive biases are fueling both extremes in the LLM discourse. Whether you're convinced we're…

SO love this post from @paulbloomatyale! Real-world effects are awesome when you can get them, but psychologists run experiments in controlled environments "for the same reason that chemists keep their test tubes clean". YES! smallpotatoes.paulbloom.net/p/its-fine-if-…

A Cognitive Scientist's guide to Machine Consciousness osf.io/s7ptu/

Just finished reading this that I strongly recommend to anyone interested in history of cognitive science (and more broadly in questions related to the advancement of science) and is curious about what happened during the “cognitive revolutions” 1/n citeseerx.ist.psu.edu/viewdoc/downlo…

Just finished reading this old-ish gem by Herbert Simon (1992) that I strongly suggest reading for all interested in the epistemological foundations of behavioural sciences. Many key ideas are extremely clearly and convincingly laid down 1/n

Will huge sample sizes (alone) solve psychology problems (as for example it is claimed they solved behavioural genetics problems)? This classic article clearly illustrate that "not quite". It start from recognising a key difference between psychology and physics 1/n

happy to see the work of my friend @MaevaLhotellier on RL range-adaptation algorithm✨

Achieving Scale-Invariant Reinforcement Learning Performance with Reward Range Normalization osf.io/bjyr9/