The general problem of visual search can be shown to be computationally intractable in a formal complexity-theoretic sense, yet visual search is widely involved in everyday perception and biological systems manage to perform it remarkably well. Complexity level analysis resolves this contradiction. Visual search can be reshaped into tractability
through approximations and by optimizing the resources devoted to visual processing. Architectural constraints can be derived to rule out a large class of potential solutions and the evidence speaks strongly against bottom-up approaches to vision. More importantly, the brain is not performing so-called general purpose vision; the vision problem the brain is solving is re-shaped using the constraints arising from the complexity analysis. In particular, the constraints argue for an attentional mechanism that exploits knowledge of the specific problem being solved. This analysis of visual search performance in terms of attentional influences on visual information processing and complexity satisfaction allows a large body of neurophysiological and psychological evidence to be tied together leading to a predictive model of vision and attention.

The Selective Tuning Model is a proposal for the explanation at the computational and behavioral levels of visual attention in humans and primates. Key characteristics of the model include a top-down coarse-to-fine winner-take-all selection process, a unique competition formulation with provable convergence properties, a task-relevant inhibitory bias mechanism, and selective inhibition in both spatial and feature dimensions for elimination of signal interference that leads to a suppressive surround for attended items. An extensive set of predictions arises many of which have now been supported by experiment. This presentation is an example of how a purely theoretical computational analysis can lead to a derivation of a model (without data-fitting or learning) that can have deep impact on brain science.