The correlational structure of brain activity dynamics, known as functional connectivity, is often taken to reveal intrinsic properties of neural processing. Functional connectivity measures observed during spontaneous activity are candidate biomarkers for neurological disorders, but the physiological bases of these measures are poorly understood. In the first part of my talk, I will discuss evidence that resting state functional connectivity of the mammalian brain derives in part from interactions with extrinsic sense organs. Central or peripheral inactivation of vibrissa-mediated input in awake rats strongly diminishes somatosensory fMRI correlations, even when no overt stimuli are present. The same neural signatures are sharply reduced in a rat model of fragile X syndrome, an autism spectrum disorder. Fragile X rats are also insensitive to vibrissa inactivation, suggesting that part of the neuroimaging phenotype in this autism model arises from abnormalities in the peripheral sensory channel. In the second part of my talk, I will discuss a molecular imaging tool our laboratory has developed to place functional connectivity findings on a physiological footing. In combination with tract-tracing viruses, a genetically encoded activity probe called NOSTIC provides a means for defining input-output relationships across the brain. We demonstrate this capability for mapping circuit function in rats and suggest its utility for analyzing functional connectivity in a variety of contexts.