Representational learning by optimization of neural manifolds in an olfactory memory network.

Olfactory discrimination training in juvenile and adult Danio rerio selectively enhances separation of neural manifolds for task-relevant odors in telencephalic area pDp, enabling manifold-capacity–based classification.

• Why it matters: Internal representations depend on neural manifold geometry, but how experience reshapes that geometry—and whether it tracks behavior—remains unclear.
• What they did: Juvenile and adult zebrafish were trained on an odor discrimination task, and telencephalic area pDp population activity was analyzed for manifold geometry changes using a manifold capacity framework.
• The result: Manifold capacity revealed multiple geometrical modifications supporting classification of task-relevant sensory information and predicted odor discrimination across individuals, linking representational geometry to behavior.

AI summary of the abstract · may contain errors · not medical advice · read the paper.