Tweet
Can a large language model be used as a "cognitive model" - meaning, a scientific artifact that helps us reason about the emergence of complex behavior and abstract representations in the human mind? My answer is YES.

Why and under what conditions? 🧵
Mentions
See All