Journal of Applied Psychology 1988, Vol. 73, No. 2, 275-280 Copyright 1988 by the American Psychological Association, Inc. 0021-9010/88/100.75 Sweet Tooth Demonstrated: Individual Differences in Preference for Both Sweet Foods and Foods Highly Sweetened M. T. Conner, A. V. Haddon, E. S. Pickering, and D. A. Booth University of Birmingham, Birmingham, England A total of 9 women and 9 men, aged 19-40 years, rated the sweetness of samples of chocolate, lime drink, and tomato soup relative to their personally ideal levels of sweetness for each food, with the conditions of each test session designed to eliminate various sources of bias. The method yielded precise estimates of the individual's most preferred concentration of sugar and of his or her tolerance of deviations from ideal. This linear precision allowed assessment of the degree to which the ideal food sugar level was a personal characteristic. We found that individuals differed significantly in peak-preferred levels of sugar, both generally across foods and with some variation relatively among foods. Also, the peak sweetness preferences for the three foods correlated significantly with several rated choices of sweet foods over nonsweet foods. This is evidence for the full construct of a sweet tooth—both a strong liking for sweet foods and a liking for strongly sweetened foods. The full construct of a sweet tooth can be considered to in- clude both a strong liking for sweet foods and a liking for strongly sweetened foods. The existence of such an individual characteristic would be of interest to food formulaters and mar- keters and to those involved in fighting dental caries and obesity. It has been suggested that sweetness preference may be the ma- jor influence on the amount of sucrose an individual consumes (Desor, Greene, & Mailer, 1975). Olson and Gemmill (1981) reported some correlation between the most preferred level of sweetness in sugar water and the choice of level of sweetness in apple juice in 4- to 5-year-olds. Pangborn and Giovanni (1984) found significant correlations in adults between the most pre- ferred level of sweetness in lemonade and both intake of sweet foods and preferences expressed for high-sugar over low-sugar foods. These studies provide only fragmentary evidence for the full construct of a sweet tooth, however. A robust and rapid method for assessing individual differ- ences in most preferred composition of foods has recently been developed (Booth, Thompson, & Shahedian, 1983; Conner, Haddon, & Booth, 1986). The method gains precision because every sweetness rating is relative to a personal ideal point, in- stead of ideal sweetness being rated only at the end of the session (Moskowitz, 1972). Precision also derives from use of a presen- tation procedure designed to reduce known biases in rating procedures (Parducci, 1963; Poulton, 1979). The method pro- duces a linear relation between the sensory ratings and ratios of Mark Conner was supported by a collaborative research studentship awarded by the UK Agricultural and Food Research Council to David Booth at the University of Birmingham and to Arthur Haddon and San- dra Pickering at Unilever Research, Colworth. Arthur Haddon is now a sensory analysis and consumer research con- sultant at 10 Parkstone Close, Bedford MK4I 8BD, England. Sandra Pickering is now with Mars Group Services, Maidenhead, England. Correspondence concerning this article should be addressed to David Booth, Department of Psychology, P.O. Box 363, University of Birming- ham, Birmingham B15 2TT, England. values of a physical parameter that has a constant Weber ratio (concentration in the case of sucrose: Schutz & Pilgrim, 1957). The determinate linearity of this psychophysical function al- lows accurate interpolation of the individual's ideal level of su- crose. It also makes it feasible to estimate the individual's toler- ance of deviations from this ideal before the food formulation is rejected. Variation in ideal sucrose concentration across assessors and foods permits assessment of one part of the construct of a sweet tooth. In the present study, the most preferred level of sugar was measured across three contrasting foods—chocolate, a lime drink, and tomato soup. Individual variation in choices be- tween sweet and nonsweet foods constitutes a second part of the construct of a sweet tooth. This was studied by questionnaire assessment of the acceptability in a specified eating context of foods or beverages differing in sweetener content in the limited number of instances in which sweet and nonsweet alternatives are widely available. The full construct would be validated to the extent that preferences for high levels of sweetness corre- lated with the strength of preference for the sweet alternative relative to the nonsweet alternative in one or more contexts. Method Subjects The assessors were 9 women and 9 men volunteers, who ranged widely in age (19-40 years), from among the students and staff of the Department of Psychology at the University of Birmingham, England. Materials Chocolates were prepared from three bases provided by Cadburys in Bournville, Birmingham. Each base contained 29.5% (by weight) har- dened palm kernel oil, 0.5% lecithin, and either 70%sugar, cocoa pow- der, or dried milk powder. Heated bases were mixed to give chocolates containing 7% cocoa, 29.5% palm oil, 0.5%lecithin, and 12 differing concentrations of sugar and dried milk, with the milk powder offset against the sugar. The final concentrations of sugar were 10.0, 12.5, 275 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.