Hubel and Wiesel’s work with the visual systems of cats presents positive evidence that pattern features are extracted in the visual cortex and are no doubt critical for pattern recognition. Researchers working on pattern recognition systems were strongly influenced by this work. Features are important. On the other hand, research of Julesz using random dot stereograms shows that visual processing can be done with featureless random dot patterns. These ideas of pattern recognition seem to conflict. A singular conclusion is that the animal visual system uses both features and the whole pattern to process visual images. An approach to pattern recognition design would benefit from analyzing both features and whole images.