To characterize restrained eating behavior, we fed pudding mixtures (preloads) varying in energy content and sweet taste to subjects before measuring their intake of a variety of appetizing foods in a laboratory setting. Twenty-four female subjects were divided into three groups: normal-weight restrained eaters (no. = 8), normal-weight non-restrained eaters (no. = 7), and over-weight restrained eaters (no. = 9). Classification of subjects was based on body mass and restrained eating score. Five preloads varying in energy content (393 or 67 kcal) and type of sweeteners (sucrose or cyclamate) were served 1/2 hour before test meals of assorted hors d'oeuvres. Subjects were not informed of the specific aim of the study, and were instructed to eat until full. Multivariate analysis of variance of repeated measures showed that food intake of all subjects was higher after eating low-energy preloads than after eating high-energy preloads, although subjects did not recognize the effect in self-reports (p less than .01). The energy intake of restrained normal-weight eaters was significantly (p less than .05) lower after eating sweet preloads than after eating non-sweet preloads. Findings suggest that a food-intake control mechanism responds to short-term energy deficits among subjects of varied body mass and restrained eating indexes. J Am Diet Assoc 90:1223-1228, 1990.