Cognition with and without Cortex: Implications from Mouse Navigation in a Novel Reconfigurable Maze

Date:

Talk abstract

A hallmark of human intelligence is its ability for rapid learning and flexible adaptation to changing environments. The human neocortex and hippocampus are considered essential neural substrates for advanced cognition, such as flexible decision-making and long-term memory formation. Accordingly, much rodent research focuses on the role of the cortex in learning and memory. To study the cognitive capacity of mice and the essential role of the cortex, we designed a novel navigation task, the “Manhattan Maze”. ​

The Manhattan Maze is easily reconfigurable and allows systematic task designs in a vast space of 2121 possible maps. We found that completely naïve wildtype mice quickly learned to navigate their first map in ~10 roundtrips, retained overnight memory of the same map, and showed accelerated learning in new configurations. We also tested acortical mice— a structural mutant born without both the hippocampus and most of the neocortex. Surprisingly, these mice were able to learn multiple maps and retained the ability to solve the same map even after a two-month break. These results suggest that the mouse cortex was not strictly required for learning and memorizing the Manhattan Maze.​

These findings raise fundamental questions about the role of the cortex in cognition. In the final part, I’ll discuss an unresolved brain paradox: the vast gap between its gigabits/s sensory inputs and 10 bits/s behavioral output. Exploring the reasons behind this paradox will shed light on the unique role of the cortex and organizing principles of the brain.


Documents

Link to the preprint “the Unbearable Slowness of Being”.

Download the talk slides here