Research & Papers

Dynamical Mechanisms for Coordinating Long-term Working Memory Based on the Precision of Spike-timing in Cortical Neurons

A legendary AI pioneer just revealed a radical new theory of cognition.

Deep Dive

Neuroscientist and AI pioneer Terrence Sejnowski has published a new perspective paper proposing a 'second tier' of neural activity for long-term working memory. The theory suggests millisecond-precision spike timing, coordinated by cortical traveling waves, creates temporary networks that can last for hours. This mechanism would operate independently of standard sensorimotor processing, potentially explaining how we hold complex thoughts and plans. The paper, spanning 37 pages, directly connects these biological mechanisms to artificial intelligence.

Why It Matters

This could fundamentally reshape how we model memory and cognition in both neuroscience and future AI systems.