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

Abstract Cognition relies on internal representations of relevant information that are organized by constraining population dynamics to activity subspaces referred to as neural manifolds. Here, to examine how manifold geometry is modified by experience, we trained juvenile and adult zebrafish in an odor discrimination task and measured population activity in telencephalic area pDp, the homolog of piriform cortex. No obvious signatures of attractor dynamics were detected; however, olfactory discrimination training selectively enhanced the separation of neural manifolds representing task-relevant odors from other representations, consistent with predictions of autoassociative network models endowed with precise synaptic balance. Analytical approaches using the framework of manifold capacity revealed multiple geometrical modifications of representational manifolds that supported the classification of task-relevant sensory information. Manifold capacity predicted odor discrimination across individuals, indicating that representational geometry is linked to behavior. Hence, pDp and possibly related recurrent networks store information in the geometry of neural manifolds, resulting in joint sensory and semantic maps that may support distributed learning processes.