Localization accuracy for a wide Gaussian bar(sigma 1 degrees) was ess
entially invariant with sampling density over a wide range. Luminance
contrast, on the contrary, had a profound effect on localization accur
acy. This difference suggests that the position is estimated from a fe
w (3-4) samples from all those available in the image. A subsequent ex
periment confirmed that limiting the display to four samples did not i
mpair localization accuracy. Computer simulations show that the result
cannot be explained by the peak or centroid rule for position. The re
sults imply that the visual system uses only a few samples to interpol
ate the luminance profiles, regardless of how many samples are availab
le in the image. (C) 1998 Elsevier Science Ltd. All rights reserved.