The purpose of this study was to examine the relationship among structural, semantic, and verbal memories of visual art. It was hypothesized that verbal contextual information, in the form of descriptive and interpretive titles, promotes varied levels of structural and semantic processing as a function of differences in abstraction, media, and prior knowledge. Findings were best explained by the sensory-semantic model of processing in predicting that the availability of contextual information at input facilitates the semantic analysis of structural information. Results indicated meanings were more accessible for processing when responding to photographs than to drawings, yet the accuracy of their interpretation was determined by the abstraction level that characterized each work of art and by the type of contextual information given during presentation. The findings further indicated that even students with prior knowledge of photography benefited from the availability of verbal contexts.