Sensory neuron responses in awake cortex can be quite variable across repeated presentations of the same stimulus. In many cortical areas, only a small fraction of neuronal activity is repeatable from trial-to-trial, raising the question of what these neurons are representing in their response, if not the stimulus? In this talk I will describe two sources of variability that account for variability in the awake cortex. In primary visual cortex (V1), I will show that nearly half of the response variability of V1 neurons can be explained by fixational eye movements. New methods of precise eye tracking can be used to control for this variability, and also allow the ability to study the effects of fixational movements in detail. In area MT, a large fraction of the variability can be explained by ongoing network inputs that are not locked in time to the visual stimulus. These inputs can be inferred from the LFP and shown to be correlated with behavior in motion discrimination tasks. The model-based descriptions of cortical processing that account for sources of variability in the awake cortex thus set the foundation for a broader understanding of visual function.