In film and still photography, depth of field control is often employed to control viewer interest in an image. This technique is also used in computer animation, but, in a medium where artists have near infinite control, must we rely on replicating photorealism? This research is a viable, non-photorealistic solution to the problem of directing viewer interest. Vision is directed by reducing superfluous visual information from parts of the image, which do not directly affect the depictive meaning of that image. This concept is applied to images and animations rendered from three-dimensional, computergenerated scenes, where detail is defined as visual information pertaining to the surface properties of a given object. A system is developed to demonstrate this concept. The system uses distance from a user-defined origin as the main mechanism to modulate detail. This solution is implemented within a modeling and shading environment to serve as a non-photorealistic, functional alternative for depth of field. This approach is conceptually based on a model of human vision, specifically, the relationship between foveal and peripheral vision, and is artistically driven by various works in the disciplines of painting and illustration, that through the careful manipulation of detail, control interest and understanding within the image. The resulting images and animations produced by this system provide viable evidence that detail modulation can be used to control effectively viewer interest in an image eliminating the need to use photographic techniques like depth of field.