Big thanks to all the co-authors. Huge shoutout to Tony Drinnenberg, Charu Ramakrishnan and @deisseroth.bsky.social as well as the team at @alleninstitute.org. This would not have been possible without their new mouse lines. Check out this preprint for more information: doi.org/10.1101/2025...

In short, we show that novelty gates rapid cortical plasticity. We think that this allows the brain to navigate a changing world without sacrificing what it’s already learned.

Together, these results outline a new framework for cortical learning: familiar representations stabilize in superficial layers, once established, they are hard to perturb. Deeper layers, on the other hand, stay more flexible to encode change.

We found a striking layer-specific rule for plasticity: Neurons in layer 2/3 of both RSC and V1 are highly plastic in novel environments but stable in familiar ones. Layer 5 neurons remain plastic regardless of familiarity, integrating new, behaviorally relevant information.

We stimulated neurons with single-cell resolution in either RSC or V1 when mice reached a predefined position in the environment. How would this stimulation affect the position-correlated activity of targeted neurons?

This approach allowed us to manipulate the activity of single neurons and watch where and when the neocortex learns new spatial information in real time.

Using all-optical “read-write” experiments—two-photon calcium imaging combined with two-photon optogenetics—we recorded and manipulated cortical neurons in mice navigating novel and familiar environments.

Congratulations! Amazing paper!!