Nathaniel D. Daw

Causal evidence that language models use confidence to drive behaviour

Metacognition—assessing the quality of one’s own cognitive performance—guides adaptive behaviour across species. Confidence signals can be extracted from language model outputs, yet a fundamental question remains: do models actually use these signals to decide whether to answer or abstain? Here we developed a four-phas...

Shifts in learning dynamics drive developmental improvements in the acquisition of structured knowledge

Sparks of In Silico Cognitive Science: Theories from Simulated Data Can Generalize to Humans

Behavioral foundation models have been proposed as stand-ins for human participants across settings, but it is unclear whether theories discovered on them generalize to humans or merely characterize the simulator. We ran the Automated Cognitive Scientist (\textsc{AutoCog}), a closed-loop discovery system in which LLM a...

The Computational Basis of Confidence in Large Language Models

A computational account of confidence in multimodal language models is provided, when answer logits behave as readouts of a latent decision variable is delineated, and statistical decision confidence is established as a unifying framework for studying confidence across biological and artificial intelligence.