Thursday, 20th March - 11am: Micha Heilbron

Talk Title: Large Language Models for cognitive neuroscience: Two success stories and a warning 

 Abstract: Large Language Models (LLMs) are not only the best AI models for linguistic tasks, but also at predicting brain responses to language, despite not being designed with the brain in mind. This surprising alignment opens two parallel research directions. The first investigates why these brain-agnostic models can capture brain responses so well, and what it teaches us about the brain. The second explores how we can actively build cognitive principles into LLMs, allowing to test cognitive hypotheses and create more brain-like models. In this talk, I present new work on both fronts. Results highlight that (1) high-level syntactic abstractions are a key driver of LLM-brain alignment; (2) human-like fleeting memory improves language learning in transformer-based language models; and (3) a recent, widely cited neural signature of next-word prediction in the brain can be fully explained by stimulus correlations, rather than next-word prediction. Together, these findings show how cognitive science can both illuminate and improve language models – and provide a cautionary tale about the difficulty but vital importance of interpretation in “NeuroAI”.