Bottom-up clues in target finding: Why a Dalmatian may be mistaken for an elephant

We provide informal psychophysical support for a strategy where bottom-up f eatures guide attention toward a target, and the top-down path interprets h ypothetical shapes at the target location-as opposed to a dominant top-down approach. In our survey, for which we used the familiar picture of a Dalma tian dog against a dappled background, (i) 75% of subjects initially found a bulging body which overlaps that of the dog, but final 'top-down' percept s were unexpected: nearly all subjects assigned an incorrect head and limbs to the body; (ii) after random rotation of texture elements overlapping co mputed features only 45% of subjects reported a bulging body, with a few ad ding limbs etc. The picture of the Dalmatian dog must therefore contain man y bottom-up features-a top-down strategy may find 'incorrect' targets at co rrect target locations. Computational support for these claims is more easi ly constructed than one may expect. We could compute at least two bottom-up features, both useful in 3-D surface interpolation from 2-D scenes, which yielded significant values at the location of the Dalmatian dog: anisotropi c texture compression and affine texture distortion cues. We therefore conc lude that the role of top-down processing is overstated in a traditional ex ample such as the Dalmatian dog picture.