"A summary-statistic representation of peripheral vision explains visual crowding" was just accepted at the Journal of Vision. Lisa Nakano and Ruth Rosenholtz are co-authors on the manuscript. The paper describes work we've done suggesting that the information that is lost in visual crowding is the by-product of the visual system's use of a texture-like pooling of visual structure in the periphery. We introduce the use of images we call "mongrels" that are synthetic textures made from crowded arrays and demonstrate that these images contain roughly the same amount of information as the orginal crowded images.