"A 'perspective-size' instruction designed to make explicit the response bias postulated to be present in size-constancy judgments of objective-size clearly produced a greater degree of over-estimation than objective-size instruction. The difference from a physically correct size-match was as great as that obtained in the opposite direction when Ss judged in terms of projective-size. Perspective-size was positively correlated with objective-size, negatively correlated with projective-size. It is argued that size-constancy represents the basic functional relationship between object-size and perceived-size, in preference to the view that perceived-size is a compromise between object-size and angular retinalimage size." (PsycINFO Database Record (c) 2006 APA, all rights reserved)