Mice in the Manhattan Maze: Rapid Learning, Flexible Routing and Generalization, With and Without Cortex

Cognitive Computational Neuroscience 2024

This CCN presentation describes the Manhattan Maze, a reconfigurable navigation task for studying cognitive flexibility in mice. Naive wildtype mice learn complex maps within two days, retain repeated maps, and accelerate learning in new configurations. Acortical mice initially take longer to solve the maze, but learn multiple maps, approach optimal performance, and solve a repeated configuration after two months.

Original event page

Presented at CCN 2024 as Poster B41 on August 8 and as a contributed talk on August 9. The updated abstract, poster, slides, and talk recording are linked below.