Effect of ideal–relative sweetness on yogurt consumption Zata Vickers a, *, Elizabeth Holton b , Jian Wang a a Minnesota-South Dakota Dairy Foods Research Center, Department of Food Science and Nutrition, University of Minnesota, 1334 Eckles Ave., St. Paul, MN 55108, USA b Frito-Lay Inc., 7701 Legacy Drive, Plano, TX 75024-4099, USA Received 25 July 2000; received in revised form 18 April 2001; accepted 4 May 2001 Abstract The objectives of this study were to determine how hedonically different sweetness levels in yogurt, determined by the ideal– relative rating method, affected taste test liking ratings and consumption in a naturalistic setting. Nineteen subjects attended a preliminary session, a taste test and three lunch tests. During the taste test, they rated yogurt with three levels of sweetness (high, optimum, and low) for six attributes. During each lunch test, they were offered a tray of nine food items, including yogurt at one of the three sweetness levels. Subjects liked the optimally sweet yogurt best in the taste test and consumed the most of it at lunch. Taste test liking ratings did not predict the amount of yogurt consumed during lunch. The lower-than-optimum sweetness level was more detrimental to taste test ratings than was the oversweetened yogurt whereas the higher-than-optimum sweetness level was more detrimental to consumption during lunch than was the undersweetened yogurt. # 2001 Elsevier Science Ltd. All rights reserved. Keywords: Optimum sweetness; Food consumption; Yogurt consumption; Hedonic scoring 1. Introduction Hedonic ratings such as those obtained using the 9- point hedonic scale (Peryam & Girardot, 1952) some- times, but not always, predict consumption. A few studies suggest that the consumption of overly sweet foods is less than would be predicted by taste test liking scores. Lucas and Bellisle (1987) showed that a higher sucrose concentration was preferred in taste-and-spit tests than in consumption tests. Drewnowski, Grinker, and Hirsch (1982) found that repeated tasting of sodas led to lower hedonic ratings, and the decreases in ratings were largest for the sweetest sodas. Vickers and Holton (1998) demonstrated that a strong-flavored iced tea was much preferred to a weak tea in a taste test, but was less preferred over a series of 20 consumption occasions. The specific sensory attribute that differed the most between those two teas was the sweetness, although other attributes differed as well. Their measures of sen- sory-specific satiety on the two tea samples showed that people drank less of the more intensely flavored tea. In contrast, other studies show similar patterns between liking ratings of sweetened foods and their consumption. Daillant and Issanchou (1991), Popper, Chaiton, and Ennis (1995), Vickers, Holton, and Wang (1998), and Zandstra, de Graaf, van Trijp, and van Severen (1999) have all shown that sugar concentrations with the highest taste test hedonic ratings, or at opti- mums determined by ideal–relative sweetness ratings, were generally the most consumed in laboratory con- sumption tests. Using ideal relative ratings of sweetness (Booth, Thompson, & Shahedian, 1983) enables one to select not only an optimum sweetness level, but to select other levels of sweetness liking equal distance above and below the optimum. Doing so allows one to examine the relative impact of oversweetening vs. undersweetening. Data of Daillant and Issanchou (1991) and Vickers et al. (1998) suggest to us that oversweetening may have less impact on decreasing consumption than under- sweetening. On the other hand, the data of Drewnowski et al. (1982), Lucas and Bellisle (1987) and Vickers and Holton (1998) suggest to us that oversweetening may decrease consumption more than undersweetening. In this study, we further investigated the relations among the levels of sweetness in yogurt determined by 0950-3293/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved. PII: S0950-3293(01)00047-7 Food Quality and Preference 12 (2001) 521–526 www.elsevier.com/locate/foodqual * Corresponding author. Tel.: +1-612-624-2257; fax: +1-612-625- 5272. E-mail address: zvickers@umn.edu (Z. Vickers).