Prior information and experience with visual stimuli enhance our ability to recognize images, but where and how does this facilitation occur in the brain? Using degraded stimuli we can tease apart the effects of bottom-up visual processes, and top-down, experience-dependent processes, as prior knowledge of the fully coherent images makes them easier to recognize. Using machine learning algorithms like Support Vector Machines (SVM's), we can then quantify the amount of information a given brain region contains about the stimulus as the subject learns the coherent image. Here we show how distinct brain regions from prefrontal cortex to V1 contain more information about degraded stimuli with prior knowledge, and that regional information in the brain persists in line with behavior. Interestingly, this effect depends critically on the complexity of the stimuli, so that prior information seems to be encoded over complex, real-world features, but not simple stimuli such as oriented gratings.