Mechanisms underlying perceptual and mnemonic interaction in the brain.

Abstract: Natural behaviors require external perceptual and internal mnemonic information to interact dynamically. For example, when navigating, we continuously exchange information about the current percept with our memory of the surrounding environment. Yet the brain systems supporting internal (default network, DN) and external (dorsal attention network, dATN) processing are traditionally viewed as antagonistic. This raises a crucial question: What neural mechanism allows the brain’s perceptual and mnemonic representations to interact? Here, I propose a mechanism facilitating this interaction: a voxel-scale retinotopic code organizes the interaction between internally- and externally-oriented brain systems. First, I describe a topographic dissociation between the brain areas supporting perception and memory of scenes, and I show that these regions interact to facilitate perception of familiar places. Second, using visual population receptive field modeling, I show that these scene-perception and memory areas’ interaction during visual perception and recall tasks is structured by a voxel-scale retinotopic code. Finally, using densely-sampled 7T fMRI data, I show that this retinotopic interaction is not unique to brain areas that process scenes: retinotopic coding is a general mechanism scaffolding the interaction between the DN, dATN, and the hippocampus. Together, these studies provide a novel framework for understanding how perceptual and mnemonic information coexist and interact in the brain and suggest that perceptually-grounded neural codes play an important role in structuring interregional interaction outside of sensory cortex.