We are investigating the visual-cortex model developed in Tomaso Poggio's group at MIT (Serre et al., 2007). In particular, we are investigating the nature of the higher-level representations constructed by this model and the model's performance on a number of visual tasks. More generally, we are interested in extending this model to be able to recognize relationships between objects and other higher-order visual concepts. To do this we will be using ideas developed in our work on perception and analogy-making in conjunction with a new version of the visual-cortex model developed by Garrett Kenyon and colleagues at Los Alamos National Laboratory.