It may be that one saliency map is computed across the
entire scene during the first fixation on that scene, or that
a new saliency map is computed in each fixation. In the
former approach, the initial map could be used to generate
an ordered set of sites that are fixated in turn. This
approach assumes that a single map is retained over
multiple fixations, an assumption that is suspect given the
evidence that metrically precise sensory information about
a scene is not retained across saccades [25,26]. The alternative
approach is to compute the saliency map anew
following each successive fixation. This approach does