Without moving your eyes, try to recognize the objects around you. Most likely, you can identify many objects that are far in your visual periphery, even though they are very blurry and poorly resolved. I believe that this form of recognition requires using knowledge of your environment and the objects it contains including the prevalence of certain objects and the spatial relations between them. This 'contextual' knowledge allows us to perform visual identification on the basis of reduced local image information, I am interested in how we obtain this kind of visual knowledge and how we use it in everyday visually guided behavior such as navigation and visual search.
A related question is how much (and what kind) of information we are able to obtain from a brief glance at a scene. Past research has suggested the answer is.... quite a lot! This ability probably relies on the capaciy to combine coarse visual information with knowledge abouth the environment you are viewing.
The research in the VML uses behavioral and computational methods in order to investigate these and other related topics in vision science.
