Our lab aims to understand the neural mechanisms of perception. Perception is an inference of the most likely objects given sensory evidence and prior expectations about the sensory world. Yet, much of sensory neuroscience has focused on how the brain faithfully represents sensory information. Our research seeks to reconstrue the goal of perception as inference rather than faithful representation, and thereby reinterpret theories of neural coding at multiple levels; from psychology-level computations to circuit-level algorithm, down to the biophysical level implementation. In this vein, a specific question that we aim to elucidate concerns the spatiotemporal scale of corticocortical communication across the sensory hierarchy. To this end, we employ an interdisciplinary approach, primarily leveraging techniques from systems neuroscience (e.g., two-photon imaging, two-photon holographic optogenetics, extracellular electrophysiology) and computational neuroscience (e.g., machine learning decoding of neural population activity, network simulation).